Comparative Genomics Reveals Accelerated Evolution in Conserved Pathways during the Diversification of Anole Lizards

Comparative Genomics Reveals Accelerated Evolution in Conserved Pathways during the... Squamates include all lizards and snakes, and display some of the most diverse and extreme morphological adaptations among vertebrates. However, compared with birds and mammals, relatively few resources exist for comparative genomic analyses of squamates, hampering efforts to understand the molecular bases of phenotypic diversification in such a speciose clade. In particular, the 400 species of anole lizard represent an extensive squamate radiation. Here, we sequence and assemble the draft genomes of three anole species—Anolis frenatus, Anolis auratus,and Anolis apletophallus—for com- parison with the available reference genome of Anolis carolinensis. Comparative analyses reveal a rapid background rate of molecular evolution consistent with a model of punctuated equilibrium, and strong purifying selection on functional ge- nomic elements in anoles. We find evidence for accelerated evolution in genes involved in behavior, sensory perception, and reproduction, as well as in genes regulating limb bud development and hindlimb specification. Morphometric analyses of anole fore and hindlimbs corroborated these findings. We detect signatures of positive selection across several genes related to the development and regulation of the forebrain, hormones, and the iguanian lizard dewlap, suggesting molecular changes underlying behavioral adaptations known to reinforce species boundaries were a key component in the diversifi- cation of anole lizards. Key words: Anolis, genomes, molecular evolution, substitution rates, adaptation, phenotypic evolution. The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non- commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Genome Biol. Evol. 10(2):489–506. doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 489 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE well as for unconstrained regions that are expected to accu- Introduction mulate mutations at the neutral rate. Thus, comparative ge- The field of amniote comparative genomics has benefited nomics creates the opportunity to understand the genetic from an influx of whole genome assemblies, due to efforts basis of species divergence, and untangle the links between across multiple collaborative research groups (Genome 10K molecular evolution and morphological change (Lanfear et al. Community of Scientists 2009; Lindblad-Toh et al. 2011; 2010). Green et al. 2014; Zhang et al. 2014). This has resulted in a One group of squamates with extensive diversification better phylogenetic sampling of genomes across the amniote within the Cenozoic is the anoles, which comprise almost tree of life and a vastly improved ability to understand the 400 species of mostly arboreal Neotropical lizards classified molecular mechanisms underlying the radiation of amniotes in the genus Anolis (Squamata: Dactyloidae). The morpholog- into terrestrial, aerial, freshwater, and marine habitats over ical, ecological, and behavioral disparity among anoles makes 300 Myr of evolution. For instance, the eutherian mammals this one of the most diverse squamates clades (Losos and share a time to most recent common ancestor (TMRCA) of Miles 2002), long serving as an ecological and evolutionary 100 Myr during the Cretaceous Period (Hedges et al. 2015), model with multiple examples of phenotypic convergence on and extensively diversified during the Cenozoic Era, which the islands of the Caribbean and in Central and South began 66 Ma. Each eutherian mammalian order is repre- America (Losos 2009). Anoles are an ideal group with which sented by at least one whole genome sequence (Speir et al. to test hypotheses about the relationship between rates of 2016), enabling investigations into the genomic changes lead- molecular and phenotypic evolution, as well as the genomic ingtobothdivergent (Kim et al. 2016) and convergent (Foote bases of many adaptive traits, as there is a robust reference et al. 2015) phenotypes. Similarly, modern birds also share a genome and annotation for the green anole (Anolis carolinen- Cretaceous TMRCA (Donoghue and Benton 2007; Hedges sis)(Alfo ¨ ldi et al. 2011; Eckalbaretal. 2013). However, despite et al. 2015), and the Avian Genome Consortium has begun their ongoing importance for the study of adaptive radiation to investigate the extensive avian evolution that occurred and species diversification, and existing genomic resources, throughout the Cenozoic at the genomic scale (Jarvis et al. there has not yet been a comparative genomic investigation 2014; Zhang et al. 2014). Although currently not as far along, that seeks to identify the loci underlying adaptive divergence numerous individual genome projects have been launched for during Anolis evolution. For instance, body size and relative squamate reptiles including iguanian lizards (Alfo¨ldi et al. limb lengths have strong adaptive significance in anoles, as 2011; Georges et al. 2015), snakes (Castoe et al. 2013; sympatric species interact differently with their environmental Vonk et al. 2013; Gilbert et al. 2014), anguimorphs (Song substrates and occupy separate areas of morphospace (Losos et al. 2015), and geckos (Liu et al. 2015). To better understand 2009). Although genetic control of limb development is not the mechanisms underlying phenotypic diversity in well studied in anoles, extensive work with model organisms squamates—one of the largest groups of vertebrates—and has identified potential regions of interest. For instance, Hox provide meaningful comparisons with the other major groups cluster paralogs control limb element patterning during devel- of amniotes (i.e., mammals and birds), more robust genomic opment (Pineault and Wellik 2014) and the T-box transcrip- sampling across the squamate lineages is sorely needed. tion factor 5 (tbx5) controls limb bud outgrowth (Rallis et al. Increased genomic sampling of speciose, phenotypically 2003). A comparative genomic approach to understanding diverse clades such as squamates can provide better estimates phenotypic diversity in anoles would provide a powerful of the number of mutational changes that occurred during way to scan for anole-specific mutations in candidate regions lineage diversification and the effects of those changes, if any, such as these, place these genetic changes in the context of on phenotypes. This can help determine whether or not un- what is expected under the background DNA substitution derlying shifts in molecular evolution are related to morpho- rate, and provide insights into how anoles have responded logical rates of change and species diversity, which has been a to selective pressures at the molecular level. subjectofdebate(Bromham et al. 2002; Davies and Here, we investigate molecular mechanisms of anole lizard Savolainen 2006). For instance, species are thought to diverge diversification using complete genome assemblies from four either by gradual change or by episodic bursts of rapid diver- species. Our goals were 3-fold: 1) to study patterns of geno- sification such as adaptive radiations (Eldredge and Gould mic divergence in anoles with respect to other amniote line- 1972). Punctuated episodes of speciation can cause dramatic ages, 2) to understand how the rate of genomic evolution changes in the rate in molecular evolution, possibly providing relates to functional genomic features and, in turn, pheno- a “mutational engine” for new phenotypes (Pagel et al. typic diversity in anoles, and 3) to identify putative adaptively 2006)—although there has been considerable debate over evolving genomic regions that contributed to the well-studied this issue (Lanfear et al. 2010). Genome annotations provide ecological and morphological changes that occurred during information about the boundaries between functional and the diversification of anoles. We focus our genome sequenc- nonfunctional loci, and DNA substitution rates can be esti- ing and assembly efforts on the following species with diverse mated for genomic regions exposed to natural selection as morphologies and ecological niches: 1) the Central American 490 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE giant anole (Anolis frenatus), which is a relatively large- using the Qiagen DNeasy Blood & Tissue Kit standard protocol. bodied species that resides high on tree trunks (Losos Specimen identifiers and geographic coordinates are listed in et al. 1991); 2) the grass anole (Anolis auratus), which supplementary table 1, Supplementary Material online. occupies grassy and bushy vegetation with narrow Paired-end DNA libraries were constructed with the TruSeq perches (Fleishman 1988); and 3) the slender anole Library Prep kit for target sizes of 180, 300, 500 bp, and 1 kb (Anolis apletophallus), which is primarily found lower on at the Translational Genomics Research Institute (TGen; tree trunks and on the ground (Ko ¨ hler and Sunyer 2008). Phoenix, Arizona). Mate pair libraries of target size 3 kb Combined with the reference genome for A. carolinensis, were constructed using the Nextera v1 Mate Pair Library Kit which resides in crowns and high trunks of trees in the (Illumina) at TGen and additional mate pair libraries of target southeastern United States, these species are distributed size 5 kb were constructed with the Nextera Mate Pair Library across the anole lizard phylogeny (Guyer and Savage kit at the University of Arizona Genetics Core (UAGC; Tucson, 1986; Poe 2004), which may have originated between Arizona). Genomic DNA was sequenced on both Illumina 120 and 45 Ma (Nicholson et al. 2012; Prates et al. HiSeq 2000 and 2500 platforms at TGen and UAGC, respec- 2015). The phylogenetic and ecological diversity of these tively. We trimmed raw reads for nucleotide biases, adaptors, species provides an ideal opportunity to study the geno- and a quality score cutoff of28 with Trimmomatic v0.33 mic underpinnings of Anolis diversification, adaptive radi- (Bolger et al. 2014). We corrected errors in the trimmed ations of tetrapods in general, and how evolution has sequences using the module SOAPec in SOAPdenovo2 (Luo shaped genomes and phenotypes during the history of et al. 2012). Overlapping reads were joined to form single-end land-dwelling vertebrates. reads using FLASH v1.2.8 (Magoc and Salzberg 2011). We compared the outputs of different genome assembly pipe- lines including ABySS versions 1.3.7 and 1.5.2 (Simpson Materials and Methods et al. 2009), SOAPdenovo2 (Luo et al. 2012), Platanus Animal Processing, Genome Sequencing, and Assembly v1.2.1 (Kajitani et al. 2014) for both contig and scaffold as- Adult A. apletophallus, A. auratus,and A. frenatus were col- sembly, and SSPACE v3.0 (Boetzer et al. 2011)for scaffold lected at field sites under permits SE/A-33-11 and SC/A-21-12 assembly. We selected assemblies with the longest contig and issued by the Republic of Panama Autoridad Nacional de scaffold N50 values for further analysis. Candidate assemblies Ambiente (ANAM; to K.K.) under Institutional Animal Care were also evaluated based on expected gene content using and Use Committee (IACUC) guidelines at the Smithsonian the Core Eukaryotic Gene Mapping Approach (CEGMA v2.5; Tropical Research Institute (Protocol 2011-0616-2014-07 to Parra et al. 2009), which aligns genome assemblies to a data- K.K.). Specimens were exported under ANAM permits SEX/A- base of 248 highly conserved eukaryotic proteins and reports 81-11 and SEX/A-71-12 (to K.K.) following regulation 50 CFR the completeness of detected orthologs, and Benchmarking Part 14 of the United States Fish & Wildlife Service. Lizards ~ Universal Single Copy Orthologs (BUSCO v1.2; Simao et al. were maintained under IACUC guidelines at Arizona State 2015), which scans assemblies for a vertebrate-specific data- University (Protocols 09-1053 R, 12-1274 R, and 15-1416 R; base of 3,023 conserved genes. supplementary table 1, Supplementary Material online). Specimens were euthanized according to ASU IACUC proto- Annotation and Analysis of Repeat Elements cols 10-1053 R and 12-1274 R (K.K.) via intracoelemic injec- tion of a solution of sodium pentobarbital/sodium phenytoin To analyze the repetitive landscapes of the three de novo and secondary confirmation by thoracotomy. Twenty-four anole lizard genomes, we first ran RepeatMasker v4.0.5 adult specimens were used for osteological analyses (http://www.repeatmasker.org)(Smit et al. 2013–2015) on (13 A. apletophallus;7 A. auratus;4 A. frenatus) together each assembly using a library of known Anolis repeat family with 10 adult A. carolinensis specimens obtained from consensus sequences from RepBase (Jurka et al. 2005). After Charles D. Sullivan Co., Inc. (Nashville, TN) and Marcus this initial masking of reference repeats (i.e., those known Cantos Reptiles (Fort Myers, FL). Given the lack of published from A. carolinensis and shared ancestrally across verte- studies comparing the limb morphology of A. apletophallus, brates), we scanned the unmasked portions of each genome A. auratus, A. frenatus,and A. carolinensis, we completed a for novel elements using RepeatModeler v1.0.8 (http://www. morphometric analysis of the forelimb and hindlimb stylopod repeatmasker.org)(Smit et al. 2008–2015; Smit et al. 2013– (humerus and femur), zeugopod (radius/ulna and tibia/fibula), 2015). This step allowed us to identify de novo repetitive and autopod (manus and pes) (supplementary methods and elements that may be unique to each species and thus would supplementary table 1, Supplementary Material online). In ad- not have been detected by RepeatMasker alone, which relies dition, for A. apletophallus, A. auratus,and A. frenatus, one on nucleotide similarity to previously archived elements. To specimen per species was selected for genomic study; we dis- estimate the amount of evolutionary divergence within repeat sected skeletal muscle tissue from A. apletophallus and liver families and characterize the repetitive landscape of each ge- from A. frenatus and A. auratus, and extracted genomic DNA nome, we generated repeat family-specific alignments and Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 491 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE calculated the average Kimura 2-parameter (K2P) distance sequence alignment (MSA) with MULTIZ v11.2 (Blanchette from consensus within each repeat family. We modified the et al. 2004)and used A. carolinensis as the reference species weight of two transition mutations as 1% of a single transi- and a guide phylogenetic tree (see Supplementary Material tion, in order to correct for high mutation rates at CpG sites. online). The MSA was filtered for the presence of aligned This was done for the A. auratus, A. apletophallus,and A. blocks from at least 26 out of the 31 species (84% complete). frenatus assemblies using the calcDivergenceFromAlign.pl tool contained within the RepeatMasker package. We Phylogenetic Analyses accessed the repeat landscape for A. carolinensis at http:// To reconstruct the amniote phylogeny and estimate neutral www.repeatmasker.org/species/anoCar.html (last accessed substitution rates across lineages, we relied on 4-fold degen- December 2016). erate (4 D) sites, which are positions in coding regions where all mutations are synonymous at the amino acid level and can Gene Annotations be used as a proxy of the neutral rate of evolution (Kumar and We created gene models for the A. frenatus, A. auratus,and Subramanian 2002). We extracted 4 D sites from the MSA, A. apletophallus genomes based on homology with protein- excluding sites from the mitochondrion, based on the A. car- coding sequences in A. carolinensis (AnoCar2.0) and the olinensis protein-coding sequence (CDS) annotations UniProt/SwissProt database (UniProt Consortium 2015), com- (Anocar2.0, Ensembl v86, last accessed December 2016) us- bined with ab initio predictions from SNAP (November 29, ing msa_view from the Phylogenetic Analysis with Space/Time 2013 release; Korf 2004) using multiple iterations of Models (PHAST) v1.3 package (Hubisz et al. 2011). MAKER v2.31.5 (Holt and Yandell 2011) for each species as- Phylogenetic reconstruction of the 31 included vertebrates sembly. The first iteration of MAKER aligned the protein using the 4D site data was performed with RAxML v8.2.3 sequences to the assembly to produce draft gene models. (Stamatakis 2014). We generated 20 maximum likelihood Ab initio gene predictors benefit from the training of their (ML) trees under the GTRCAT substitution model, setting Hidden Markov Models (HMM), and we trained SNAP by run- Latimeria, a lobe-finned fish, as the outgroup, and conducted ning MAKER a second time with SNAP using the species- 500 bootstrap replicates, placing the bipartitions from the specific HMMs generated from the CEGMA analysis described bootstrap analysis on the ML tree with the highest likelihood earlier (for A. apletophallus, we used the CEGMA gene mod- to assess node support. els from A. auratus, which were more complete—see Results). Using the gene models from this output, we then generated Rates of Molecular Evolution an improved HMM for SNAP in a third MAKER iteration. A We estimated absolute rates of molecular evolution in terms fourth and final run of MAKER was then performed, which of substitutions per site per million years across the best ML incorporated the final SNAP HMM and the aligned protein tree, and estimated the divergence times of the major amni- data, resulting in the final set of gene model predictions. For ote lineages, including the TMRCA of anole lizards, using the each species, the final genes were functionally annotated semiparametric method based on penalized likelihood (PL) in with two methods: 1) a BLASTp of MAKER2-predicted pro- r8s v1.8 (Sanderson 2002, 2003). We used two sets of criteria teins to known proteins in UniProt, and 2) the scanning of to constrain minimum, maximum, or fixed node ages. The MAKER-predicted proteins for the presence of functional first set of criteria was based on constraints from the paleon- domains using InterProScan 5 (Jones et al. 2014; Mitchell tological literature; specifically, the minimum age of Diapsida et al. 2015). Functional annotations were incorporated using (i.e., the TMRCA of chicken and Anolis) was set at 255.9 Myr, scripts provided with the MAKER2 package. the minimum age of Archosauria (i.e., the TMRCA of chicken and alligator) was set at 247.1 Myr, the minimum and max- Whole Genome Alignments imum ages for Mammalia (i.e., the TMRCA of platypus and As the basis for our comparative genomic study of amniotes, human) were set at 164.9–201.5 Myr, and the minimum and we generated a whole genome alignment of 31 sarcoptery- maximum ages for Neognathae (i.e., the TMRCA of chicken gian vertebrates, featuring representatives of the major and zebra finch) were set at 66–86.8 Myr (Benton et al. groups of amniotes (mammals, squamates, turtles, crocodili- 2015). The second analysis used comparatively fewer con- ans, and birds) and ten available squamate genomes including straints, using the node age estimate for Archosauria of the three de novo anole lizard genomes (supplementary table 247–250 Myr obtained from TimeTree (www.timetree.org; 2, Supplementary Material online). First, we generated pair- Hedges et al. 2015). We set the minimum and maximum wise syntenic alignments of each species’ genome to the ages for Crocodilia at 77.8–83.6 Myr (Green et al. 2014) A. carolinensis genome (AnoCar2.0) using LASTZ v1.02 and fixed the root age (i.e., the TMRCA of coelacanth and (Harris 2007), followed by chaining to form gapless blocks human) at 415 Myr in both analyses (Donoghue and Benton and netting to rank the highest scoring chains (Kent et al. 2007). The PL method estimates a different substitution rate 2003). We used the pairwise nets to construct a multiple on each branch and implements a penalty when rates differ 492 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE greatly between branches, quantified by a smoothing param- Detecting Nonneutral Substitution Rates and Identification eter for which larger values indicate a clock-like model of Anole-Accelerated Genomic Regions (Sanderson 2002). We used cross-validation to optimize In order to identify genomic regions underlying phenotypic this smoothing parameter, allowing values to range on a variation in anoles, we aimed to detect loci in our aligned log scale starting from 10 with the exponent increasing sequences that depart from neutral expectations, using 0.3 for a total of ten steps, and reran the analysis with the phyloP (Pollard et al. 2010) to estimate nonneutral substitu- optimal value. We used the gradient check implemented in tion rates both across the amniote phylogeny and specifically r8s to ensure the signs of any active constraints were cor- across the anole lineage. PhyloP computes 1) a null distribu- rect (i.e., negative if a minimum constraint was used). tion for the total number of substitutions at a locus given a As the evolution of anole lizards constitutes several radi- nonconserved (neutral) phylogenetic model, 2) an estimate of ations resulting in 400 species across two continents and the observed number of substitutions in the alignment, and 3) several islands, we predicted that estimated substitution the P value from a comparison of this estimate to the null rates in this lineage should be faster than the phylogenetic distribution calculated with a likelihood ratio test (LRT). First, average for amniotes and that this could be explained by we estimated the neutral model by fitting a time-reversible either punctuated evolution (Pagel et al. 2006) or ecolog- substitution model (REV) to the best ML phylogeny obtained ical opportunity (Mahler et al. 2010). In a punctuated sce- from the 4D site data and transforming branch lengths in nario, the total genetic distance between the root and tip terms of substitutions per site using phyloFit in PHAST. of a phylogeny (i.e., the path length) will be associated with Second, we used phyloP to produce a conservation- the number of speciation events in a lineage (Webster et al. acceleration (CONACC) score for every site in the alignment, 2003; Pagel et al. 2006). Alternatively, in the case of eco- which is a two-sided LRT-based P value converted into either a logical opportunity, the rate of evolution is correlated with negative value to indicate accelerated evolution or positive the rate of speciation (Mahler et al. 2010). In either case, value to indicate conservation. We analyzed the distributions path lengths and speciation events inferred from molecular of CONACC scores for different annotation features: CDS, data will be positively correlated (Rabosky 2012). To test 0 0 introns, 5 -UTR, 3 -UTR, intergenic regions, and ancestral this prediction, we first searched for evidence of punctu- repeats (AR), under the expectation that AR should evolve ated molecular evolution across the best ML phylogeny under relatively weaker purifying selection (Mouse Genome using a phylogenetic generalized least squares (PGLS) re- Sequencing Consortium et al. 2002) and therefore possess gression in BayesTraits v3.0 (Pagel et al. 2004), which incor- lower CONACC scores. We conducted two types of tests: porates a species phylogeny and weights data points 1) an “all branches” test that examines rate variation across according to shared ancestry. We removed the non- every branch in the phylogeny, and 2) multiple “subtree” amniote outgroup and applied a continuous model to the tests which separately examine rate variation within a partic- phylogenetic path length (i.e., the total number of substi- ular clade relative to the rest of the phylogeny. As suggested tutions per site) and the number of nodes (i.e., speciation by Pollard et al. (2010), we computed CONACC scores for events) leading to each of the 30 terminal taxa as depen- subtrees based on 10-bp windows. We compared subtree dent and independent variables, respectively. We then es- test results for anoles and snakes. As a test of the null hypoth- timated 1) the proportion of evolution across the tree esis that substitution rates between a subtree and the rest of attributed to speciation and 2) the deviation from the mo- the phylogeny are equal, we compared the subtree CONACC lecular clock due to punctuational events, using the test for score distributions for each annotation feature using the punctuational evolution available at http://www.evolution. Mann–Whitney U-test inRv3.3.0(R Core Team 2016). rdg.ac.uk/pe/index.html (last accessed April 2017), testing Finally, to determine which protein-coding genes contain for a node-density artifact which can lead to the underes- anole-accelerated regions, we used phyloP to compute one- timation of branch lengths due to regions of the phylogeny sided P values to indicate accelerated evolution (ACC). We with fewer taxa (Webster et al. 2003; Venditti et al. 2006). then collected ACC regions with a P value of 0.001 which Finally, we generated a continuous regression model using overlap with Ensembl v86 AnoCar2.0 whole gene annota- the path lengths and number of speciation events for each tions using bedtools intersect (Quinlan and Hall 2010). We taxon, and used it to predict the path lengths for A. frena- tested for statistical enrichment of overrepresented GO terms tus, A. carolinensis, A. auratus, and A. apletophallus using associated with the genes overlapping accelerated regions BayesTraits. All MCMC analyses were run for an initial using the PANTHER overrepresentation analysis tool available 10,000 generations for burn-in, followed by 1,000,000 on the Gene Ontology Consortium website (GO Ontology iterations with sampling every 1,000 generations. We database Released April 24, 2017, www.geneontology.org, used Tracer v1.6 (http://tree.bio.ed.ac.uk/software/tracer/; last accessed May 2017; Gene Ontology Consortium 2015), last accessed December 2016) to monitor MCMC conver- applying a Bonferroni correction to account for multiple hy- gence, including non auto-correlation, by ensuring an effective sample size (ESS) of 200 for all parameters. pothesis testing. Genes overlapping accelerated regions were Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 493 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE compared with the reference set of 18,527 A. carolinensis concatenated three sets of 30 randomly selected genes and genes. We also collected overrepresented GO terms from estimated d /d (the parameter x) on all terminal branches of N S ACC analyses for the snakes subtree in order to compare the sauropsid phylogeny using branch tests in codeml (mod- functional category enrichment between squamate clades. el¼ 2, Nsites¼ 0). To identify specific genes undergoing pos- itive selection in each anole lizard genome, we performed the branch-site test for each gene in codeml (model¼ 2, Selection on Protein-Coding Genes Nsites¼ 2). The gene phylogeny was inferred independently To more directly measure selective pressures acting on protein for each locus using PhyML v3.0 (Guindon et al. 2010). The coding genes during Anolis diversification and identify genes custom pipeline used to perform these analyses is available at related to ecologically relevant adaptations, we estimated the https://github.com/WilsonSayresLab/AlignmentProcessor.We ratio of nonsynonymous to synonymous substitutions (d /d ) performed independent tests with each anole lizard species as N S in a series of analyses. Genes with pairwise d /d > 1 harbor the foreground branch, comparing the likelihood of every N S an excess of nonsynonymous and potentially functionally rel- branch-site model to a null model in which x was fixed at evant changes at the amino acid level and are candidates for 1, and the likelihood ratio test (LRT) statistic (2DLnl) to the chi- inferring positive selection (Fay and Wu 2003). First, we square distribution with a degree of freedom equal to 1. downloaded Ensembl gene annotations for the A. carolinensis Genes with a P value< 0.01 after a Bonferroni adjustment genome (AnoCar2.0) from the UCSC Genome Browser (Speir for multiple testing and a x> 1 were identified as positively et al. 2016) and assembled exons by gene based on the pair- selected. When analyzing sequence data including species wise syntenic nets between A. carolinensis and A. apletophal- which diverged over 200 Ma, saturation at synonymous sites lus, A. auratus,and A. frenatus, respectively (see genome is a potential concern; however, studies have shown that the alignment methods above), using the “stitch gene blocks” branch-site test is quite robust to this issue. In fact, in this tool in Galaxy (Blankenberg et al. 2011). Orthologous gene scenario false negatives are far more likely than false positives alignments of each species to A. carolinensis were filtered for (Gharib and Robinson-Rechavi 2013), thus making our anal- quality through a custom pipeline (https://github.com/ ysis more conservative. To link positively selected genes in WilsonSayresLab/Anole_expression/tree/master/Scripts), each species to potential phenotypes, we examined statisti- which deleted frame-shift mutations and replaced internal cally enriched GO categories as above, and tested for enrich- stop codons with gaps. We calculated substitution rates at ment of mouse phenotypes using the gene IDs of the human nonsynonymous (d ) and synonymous sites (d )and esti- orthologs in MamPhEA (Weng and Liao 2010). N S mated d /d for eachde novospecies compared with A. In anoles, limb morphology is related to locomotor differ- N S carolinensis using KaKs_calculator v2.0 (Wang et al. 2010). ences and ecological niche adaptation among sympatric spe- We applied a Bonferroni adjustment for multiple hypothesis cies (Losos 2009); therefore, we predict strong selective testing to identify positively selected genes with highly signif- pressure on limb structure and function during anole lizard icant d /d > 1. To link genes with d /d > 1 to potential diversification will result in larger d /d values for limb-related N S N S N S phenotypes, we used the Ensembl gene IDs to collect GO genes. We selected 30 protein-coding genes annotated in GO terms in Biomart (Kinsella et al. 2011). We also examined processes for limb development (GO: 0060173), limb mor- the distribution of d /d for genes belonging to GO categories phogenesis (GO: 0035108), and limb bud formation (GO: N S expected apriori to be relevant to anole lizard diversification 0060174). Orthologous sequences for each gene were col- based on the literature (Losos 2009): Reproduction (GO: lected from human (Homo sapiens), mouse (Mus musculus), 0000003), pigmentation (GO: 0043473), somitogenesis cow (Bos taurus), African elephant (Loxodonta africanus), (GO: 0001756), visual perception (GO: 0007601), perception American opossum (Monodelphis domesticus), western of smell (GO: 0007608), and limb development (GO: clawed frog (Xenopus tropicalis), chicken (Gallus gallus), 0060173) (Gene Ontology Consortium 2015). wild turkey (Meleagris gallopavo), mallard duck (Anas platyr- To analyze substitution rates in protein-coding genes hyncos), zebra finch (Taenopygia guttata), medium ground across sauropsids and identify positively selected genes in finch (Geospiza fortis), budgerigar (Melopsittacus undulatus), the four anole lizard genomes, we tested codon-based mod- painted turtle (Chrysemys picta), Chinese softshell turtle els of molecular evolution with codeml in PAML v4.9 (Yang (Pelodiscus sinensis), American alligator (Alligator mississip- 2007). We filtered the MSA to include only the 22 sauropsids piensis), Burmese python (Python bivittatus), and green anole (including representative squamates, crocodilians, birds, and (Anolis carolinensis). For the three de novo anole lizard turtles), stitched exons for each gene with Galaxy based on genomes, we generated predicted transcript sequence using the Ensembl v86 AnoCar2.0 annotations and quality filtered genomic scaffold sequences aligning to A. carolinensis exons for frame shifts and stop codons as above, and retained only by BLASTn and genome annotation data. Homologous gene gene alignments for which all species were present. To deter- sequences were aligned by codon using MUSCLE (Edgar mine if genome-wide d /d in protein-coding genes is related 2004) and edited by eye to remove ambiguously aligning N S to the background rate of molecular evolution in anoles, we regions and stop codons, with care taken to maintain the 494 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE integrity of the amino acid sequence by frequent BLAST A. frenatus, A. auratus,and A. apletophallus, respectively. All searches to NCBI. We analyzed the codon alignments of three de novo draft assemblies were 2Gb inlength, compa- limb genes using branch tests in codeml. For each gene, we rablein sizeto the estimated 2.2-Gb A. carolinensis genome assumed the species tree topology and compared three mod- and 1.78-Gb AnoCar2.0 genome assembly (Alfo ¨ ldi et al. els which allow x to vary along different branches in the 2011). Genome assembly statistics are summarized in table 2. phylogeny: 1) the null model (M0; model¼ 0, Nsites¼ 0) in The four anole genome assemblies differed in terms of which a single x is estimated for all branches on the phylog- estimated gene content, with the assemblies of A. frenatus eny, 2) an alternative model (M2a; model¼ 2, Nsites¼ 0) that and A. auratus approaching similar completeness to that of allows one x for the Anolis branches and another for all other A. carolinensis (table 2 and fig. 1A). In the A. frenatus assem- branches, and 3) a second alternative model (M2b; mod- bly, 86% of core eukaryotic genes (CEGs) were at least par- el¼ 2, Nsites¼ 0) that allows x to vary among Anolis, python, tially represented, whereas A. auratus had 97% at least alligator, turtles, birds, eutherian mammals, and all remaining partially represented CEGs; and both are similar to A. caroli- branches. We compared each alternative model to the null nensis (AnoCar2.0), which contains 91% of CEGs. In the case model by comparing the LRT statistic to the chi square distri- of A. apletophallus, 65% of CEGs were represented. BUSCO bution given the degrees of freedom determined by the dif- analyses estimate slightly lower gene representation than ference in the number of model parameters. Where the null CEGMA across all anole genomes, including 71% for was rejected, we then compared the two alternative models A. frenatus,85% for A. auratus, and 47% for A. apletophal- using the LRT. To compare relative selective pressure on limb lus, compared with 88% BUSCO representation in A. caroli- development genes among taxa, we observed the plus or nensis. The relatively low estimated gene content in the minus differential in x between Anolis and eutherian mam- A. apletophallus assembly likely reflects high fragmentation mals and birds, respectively. due to the lower sequencing coverage and high heterozygos- ity in the sequenced A. apletophallus specimen. Indeed, anal- ysis of the kmer frequency distribution generated from the Conservation of Regulatory Elements A. apletophallus Illumina reads was highly bimodal (supple- Much of the phenotypic diversity within and between species mentary fig. 1, Supplementary Material online) and suggestive is due to cis-regulatory sequences that control gene expres- of a high proportion of heterozygous sites, which can hinder sion, including enhancers (Wittkopp and Kalay 2012). In order genome assembly efforts (Kajitani et al. 2014). Analysis of the to determine if mutations in enhancer elements affected evo- distribution of GC content across the three de novo anole lutionary change during the evolution of anole lizards, we lizard genomes reveals similar distributions which lie within annotated mouse enhancer regions identified from brain, the peaks of both A. carolinensis and chicken (Gallus gallus, liver, limb, and heart tissues and compared their conservation fig. 1B), albeit with potentially more GC-bias in A. frenatus. in A. carolinensis, A. auratus, A. apletophallus,and A. frenatus Using both homology-based and ab initio methods for the (supplementary methods, Supplementary Material online). final gene annotations for each de novo anole lizard assembly, we approached the number of predicted protein-coding genes in the A. carolinensis genome (22,962 genes predicted; Results Eckalbar et al. 2013, table 2). Large proportions of annotated Genome Assemblies and Gene Annotations genes consist of BLASTp matches to the SwissProt-UniProtKB We carried out Illumina platform-based whole genome se- database and contain known protein domains; thus, all three quencing of three anole species representing both Dactyloa de novo anole genomes presented here contain genes with and Norops clades, using combinations of high coverage clear orthology in other species. The A. auratus assembly con- paired-end sequencing plus “jumping” mate pair libraries of tains 19,838 annotated protein-coding genes, with 87% varying sizes (table 1). We sequenced the Dactyloa group (17,210 genes) having significant BLASTp matches, and species A. frenatus at 116 coverage, and the two Norops 88% (17,394 genes) with known protein domains. We an- group species, A. auratus and A. apletophallus, at 141 cov- notated 19,923 protein-coding genes in the A. frenatus as- erage and 56 coverage, respectively. The lower coverage for sembly; 89% (17,771 genes) of these contain a significant A. apletophallus is attributable to technical failures in the se- BLASTp match and 92% (18,353 genes) contain protein quencing of some of the 300-bp insert libraries, the down- domains. Meanwhile, we were able to annotate 12,816 stream effects of which we attempted to ameliorate by genes for A. apletophallus; 82% of these produced both sig- adding additional coverage using 500-bp insert libraries. nificant BLASTp hits (10,483 genes) and protein domains Anolis frenatus and A. auratus contigs were assembled with (10,534 genes). ABySS v1.3.7 and scaffolded with SOAPdenovo2; A. apleto- Repetitive elements make up similar proportions of each phallus contigs and scaffolds were assembled using ABySS sequenced Anolis lizard genome (A. carolinensis¼ 33%, (supplementary tables 3–5, Supplementary Material online). A. frenatus¼ 37%, A. auratus¼ 39%, A. apletophallus¼ 29%; Scaffold N50 statistics calculated were 37, 49, and 9.5 kb for supplementary tables 6–9, Supplementary Material online). Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 495 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE Table 1 Whole Genome Sequencing Data Obtained from Three Anole Lizard Species Libraries Anolis frenatus Anolis auratus Anolis apletophallus Total Data (Mb) Coverage Total Data (Mb) Coverage Total Data (Mb) Coverage 200-bp paired-end 1,400 70 1,058 43 359 17 300-bp paired-end 502 22 1,454 63 328 11 500-bp paired-end – – – – 154 7 1-kb paired-end 94 9 515 23 244 11 3-kb mate pair 628 11 452 8 404 7 5-kb mate pair 90 4 81 4 85 4 Total 2,714 116 3,560 141 1,574 56 Table 2 Descriptive Statistics for Four Anole Lizard Genome Assemblies Included in This Study Anolis carolinensis (AnoCar2.0) Anolis frenatus (Afre1.0) Anolis auratus (Aaur1.0) Anolis apletophallus (Aapl1.0) Contigs Scaffolds Contigs Scaffolds Contigs Scaffolds Contigs Scaffolds N50 (bp) 79,867 150,641,573 23,240 36,910 19,858 48,995 2,534 9,520 L50 (Number) 6,217 5 21,942 14,499 26,569 11,107 206,073 53,667 Longest (bp) 582,046 263,920,458 448,330 649,467 304,182 563,800 110,998 217,008 Total length (Gb) 1.70 1.79 1.92 2.03 1.94 2.02 1.89 2.18 Number 1 kb 41,987 6,457 176,895 649,467 230,878 141,345 659,833 454,194 Number 100 kb 4,247 852 809 2,134 392 3,038 1 103 Percent gaps 5% 6% 4% 14% BUSCOs C: 73%[D: 1.4%], C: 50%[D: 0.7%], C: 68%,[D: 1.1%], C: 28%[D: 2.1%], F: 15%, M: 12% F: 20%, M: 29% F: 16%, M: 15% F: 19%, M: 53% Number of 22,962 19,923 19,838 12,816 annotated genes Percent repeats 33% 37% 39% 29% NOTE.—BUSCO, Benchmarking Single Copy Orthologs; C, Complete; D, Duplicated; F, Fragmented; M, Missing. We found evidence for the differential accumulation and signatures of ancient L1 and L2 retrotransposon expan- loss of transposable elements across anole lizard genomes, sions. The lack of SINEs found at <5% K2P divergence in such as R2 non-LTR retrotransposons which occupy be- A. frenatus, coupled with the low number of copies we tween 0.92% and 1.37% of the genomes of A. carolinen- detected in the genome, suggest this element family may sis, A. auratus, and A. apletophallus genomes yet are be near extinction in the Dactyloan lineage of anoles. scarcely present in A. frenatus at 0.04%. Other differences lie in the abundance of SINEs, which are three times more Phylogenetic Results and Support for a Cenozoic Origin of abundant in A. carolinensis than in A. frenatus. The starkest Anoles difference between the repetitive landscapes of the four Anolis genomes is the evidence of recent activity coupled We reconstructed the ML phylogeny of 31 sarcopterygian with the loss of ancient elements in A. carolinensis, which vertebrates, including four anole lizards, using a DNA align- stands out in comparison with the other three assemblies ment of 4 D sites that contained 763,131 distinct patterns (supplementary fig. 2, Supplementary Material online). For with 5.17% gaps or undetermined characters. All branches instance, the divergence profile of interspersed repeats in of the highest scoring ML tree had full support from the boot- A. carolinensis is characterized by a relative lack of trans- strap replicates, and the topology was highly congruent with posable elements beyond 5–10% K2P divergence from other phylogenomic studies of amniotes (fig. 2 and their family consensus and is indicative of their recent in- supplementary fig. 3 and table 10, Supplementary Material sertion in the genome. In contrast, the A. auratus, A. aple- online). This includes several previously controversial nodes in tophallus, and A. frenatus assemblies harbor a greater the amniote phylogeny, most notably a monophyletic proportion of their elements far beyond this level of diver- Atlantogenata (including elephant and armadillo) in the mam- gence (i.e., up to 45% K2P) and contain, for example, the malian subtree (Tarver et al. 2016) and full support for a 496 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE (A) (B) 0.15 Complete Fragmented Anolis carolinensis Anolis frenatus BUSCO CEGMA Anolis auratus ● Anolis apletophallus 0.10 Gallus gallus 0.05 ● ●●● ●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●● 0.00 20 40 60 80 GC Percentage FIG.1.—(A) Proportion of conserved orthologs in either complete or fragmented form detected in four anole lizard genomes. Proportions are reported out of 3,023 BUSCO (Benchmarking Universal Single Copy Orthologs) and 248 CEGMA (Conserved Eukaryotic Gene Mapping Approach) orthologs. (B) Distribution of GC content in the genomes of four anole lizards and chicken, calculated in 5-kb windows. turtle-archosaur clade (Archelosauria) (Crawford et al. 2012, (Glor et al. 2005), and the Central American A. auratus and 2015; Green et al. 2014). Within the included squamates, the A. apletophallus was estimated to be 35 Myr, suggesting that gecko forms the basal lineage, snakes are monophyletic and the recolonization of the mainland by Caribbean anoles of the the four anole lizards and Pogona support a monophyletic Norops lineage (Nicholson et al. 2005) occurred after the late Iguania, consistent with several studies (Wiens et al. 2012; Eocene. The r8s analysis using fewer constraints resulted in Pyron et al. 2013). We obtained full support for the some slightly different node age estimates, particularly in Dactyloan A. frenatus belonging to the basal lineage of mammals; however the TMRCA for anoles did not differ sig- anoles, followed by the divergence of A. carolinensis and nificantly. We also used a read-mapping approach to recon- monophyletic Norops species A. auratus and A. apletophallus, struct the mitochondrial genomes of A. frenatus and consistent with other phylogenetic studies of anoles (Guyer A. auratus, obtained orthologous mitogenomic data from and Savage 1986; Nicholson et al. 2012; Del Rosario iguanian lizards available on NCBI (supplementary table 11, Castaneda and De Queiroz 2013; Prates et al. 2015). Supplementary Material Online), and used an internal fossil The fossil-based node constraints we implemented in r8s calibration to estimate divergence times with a partitioned passed all gradient checks, and our estimates for the origin of data set using BEAST v1.8 (Drummond et al. 2012)(supple- amniotes at 303 Ma, the origin of sauropsids at 278 Ma, and mentary methods, Supplementary Material Online). This mi- the origin of placental mammals at 99 Ma are consistent with tochondrial analysis was done prior to obtaining the sequence previously published estimates (Donoghue and Benton 2007; data for A. apletophallus and thus this species was not in- supplementary table 10, Supplementary Material online). cluded. The results from the BEAST analysis were consistent Regarding the lineage leading to squamates including anoles, with those from the 4D site analysis, placing the 95% highest we estimated that the gecko lineage split from other squa- posterior density (HPD) of the TMRCA for iguanian lizards mates 175 Ma, and the TMRCA for iguanian lizards was between 195 and 113 Myr, the TMRCA of A. frenatus and 136 Myr. The TMRCA of the anoles was estimated to be A. carolinensis between 67 and 41 Myr (mean 53 Myr), and 45 Myr, suggesting a Cenozoic origin, specifically during the the TMRCA of A. carolinensis and A. auratus between 52 and Eocene which lasted between 56 and 34 Ma. The TMRCA of 27 Myr (mean 40 Myr) (supplementary fig. 4, Supplementary Anolis carolinensis, which is native to southeastern Material Online). Thus, both 4 D sites and mitogenomic data North America and originated from a Cuban radiation support a hypothesis in which anole lizard evolution occurred Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 497 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Anolis carolinensis Anolis frenatus Anolis auratus Anolis apletophallus Anolis carolinensis Anolis frenatus Anolis auratus Anolis apletophallus Proportion of Conserved Orthologs Detected Proportion of 5kb Regions Tollis et al. GBE Anolis apletophallus (A) Anolis auratus 0.0020 Anolis carolinensis Anolis frenatus Pogona vitticeps Python bivittatus Boa constricor 0.0018 Ophiophagus hannah Crotalus mitchellii Gekko japonicus Pelodiscus sinensis Chrysemys picta bellii 0.0016 ● Alligator mississippiensis Crocodylus porosus Gavialis gangeticus Struthio camelus 0.0014 Anas platyrhynchos Meleagris gallopavo Gallus gallus Melopsittacus undulatus Taenopygia guttata Geospiza fortis Ornithorhynchus anatinus Monodelphis domestica (B) Dasypus novemcinctus Loxodonta africana Bos taurus Mus musculus Canis lupus familiaris Mus musculus Homo sapiens 0.8 Latimeria chalumnae ● Anolis apletophallus Anolis auratus ●● ●● ●● 0.2 ● ● ● Anolis frenatus FIG.2.—Phylogeny of 31 vertebrates from 763,131 4-fold degenerate Anolis carolinensis ● 0.6 sites reconstructed using maximum likelihood in RAxML v8.2.3 (Stamatakis 2014). All branches received full support from 500 bootstrap replicates. Representative images of species are not drawn to scale. ● 0.4 within the Cenozoic Era, and that the Norops colonization of ● ●● Central and South America began after the late Eocene. ● Chrysemys picta bellii Speciation Events Accelerated Molecular Evolution in Anoles A comparison of the branch lengths of the phylogeny in FIG.3.—(A) Comparison of the estimated DNA substitution rates (in terms of substitutions per site per million years) on terminal and internal figure 2 suggests rapid evolution at 4 D sites along the squa- branches of the amniote phylogeny. (B) Scatterplot showing the positive mate branches, including the ancestral anole branch, as well relationship (R ¼ 0.17, P¼ 0.02) between the total phylogenetic path as the common ancestor branches of birds and mammals, length (i.e., the sum of substitutions per site from the root to the terminal respectively. Using the PL method, estimated rates of neutral taxon) and the number of speciation events (i.e., the number of nodes divergence at 4 D sites varied greatly across amniotes, and leading to the terminal taxon). Values for the four anole lizards studied differed significantly among phylogenetic groups (fig. 3A). here and two outlier taxa (house mouse, Mus musculus and western The optimal value for the smoothing parameter was 1, sug- painted turtle, Chrysemys picta bellii) are shown for comparison. Color gesting significant deviation from the molecular clock. We legend is as follows: mammals¼ orange, squamates¼ green, tur- observed that, among the sauropsids included in this study, tles¼ blue, crocodilians¼ violet, birds¼ gray. anoles and snakes have diverged at the fastest evolutionary since the split from their shared archosaurian ancestor rates in terms of DNA substitutions per site per million years. 247 Ma. Finally, DNA substitution rates at 4 D sites in mam- Consistent with previous studies, we estimated that the turtle mals surpass those of all other amniotes. lineage evolved at a far slower rate (Shaffer et al. 2013; Wang We found support for punctuated evolution across et al. 2013), as has the crocodilian lineage (Green et al. 2014). amniotes at 4 D sites, with the slope of the regression be- We also found that birds have departed significantly from tween path lengths and speciation events significantly >0 crocodilians in terms of DNA substitution rates at 4 D sites 498 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Anoles Snakes Other Squamates Turtles Crocodilians Birds Mammals Phylogenetic Path Length DNA Substitution Rate Accelerated Evolution during Anole Lizard Diversification GBE (R ¼ 0.171, P< 0.02, 95% HPD¼ 0.170–0.174, fig. 3B). In was diencephalon development (GO: 0021536, 4.37-fold en- addition, 26% of evolution across the amniote phylogeny was richment, P¼ 4.26 10 ), which was represented by 14 attributed to speciation events, with 17% of deviation from genes (vs. 3.2 expected), followed by endocrine system devel- the molecular clock caused by punctuational evolution. We opment (GO: 0035270, 3.69-fold enrichment, P¼ 4.20 found no evidence for the node-density artifact. Consistent 10 ), whichwas representedby 17genes (vs. 4.60 with the PL analysis, we estimated from the regression that the expected). We detected 2,810 genes overlapping accelerated turtle, crocodilian, and avian lineages have evolved at compar- regions unique to snakes; GO terms with the highest fold atively much slower rates that are below the trend line for enrichment in snakes were artery morphogenesis (GO: punctuated evolution. In contrast, mammals and squamates 0048844, 3-fold enrichment, 27 observed vs. 9 expected, were above this trend line. Convergence was reached in all P¼ 5.78 10 ) and artery development (GO: 0035904, 3- MCMC analyses, with ESS for all parameters 200. The re- fold enrichment, 23 observed vs. 7.8 expected, gression model accurately predicted the estimated path P¼ 4.51 10 ). lengths for A. frenatus (median 0.69 substitutions per site both predicted and observed, 95% HPD 0.50–0.90), A. caro- Accelerated Evolution and Positive Selection Acting on linensis (median 0.72 predicted vs. 0.71 observed, 95% HPD Protein-Coding Genes during Anolis Diversification 0.53–0.93), A. auratus (median 0.75 predicted vs. 0.73 ob- served, 95% HPD 0.53–0.96), and A. apletophallus (median We computed d /d through pairwise comparisons of N S 0.76 predicted vs. 0.73 observed, 95% HPD 0.57–0.99). 10,991, 13,783, and 12,246 filtered orthologous genes be- These results suggest that rapid molecular evolution in anoles tween A. carolinensis and A. frenatus, A. auratus,and is linked to rapid speciation rates during lineage diversification. A. apletophallus, respectively. Overall, we found high levels of protein-coding sequence conservation, with 88–93% of pairwise d /d estimates for genes in each comparison <0.5 N S Pervasive Conservation at the DNA Level for Functional (supplementary tables 12–14, Supplementary Material on- Elements in Anoles line), suggesting that the evolution for a majority of genes We used CONACC scores from phyloP to assess departures can be characterized by strong purifying selection (supple- 0 0 from neutrality in CDS, 5 -UTR, 3 -UTR, introns, intergenic mentary fig. 5, Supplementary Material online). Pairwise d / regions, and AR, first across all branches of the vertebrate d was >1 for 53 genes in the A. carolinensis–A. frenatus phylogeny and then in subtree analyses of anoles and snakes comparison, including genes associated with specific GO (fig. 4). The distribution of CONACC scores showed clear terms for biological processes such as locomotion (GO: differences by genomic feature in the all branches test. For 0040011), cellular response to insulin-like growth factor stim- instance, CDS sequences were by far the most conserved of ulus (GO: 1990314), execution phase of apoptosis (GO: the analyzed functional elements, with the highest levels of 0097194), and regulation of reactive oxygen species (GO: acceleration in AR, consistent with expectations that evolution 1903426), possibly reflecting adaptive differences in adult in coding regions is constrained with respect to nonfunctional body size, metabolic rate, and locomotor performance be- repetitive regions. In both subtree analyses, CONACC scores tween the two species. The 30 accelerated genes in the for AR were significantly different from other annotation fea- A. carolinensis–A. auratus comparison were associated with tures (Mann–Whitney U-test, P¼ 0). CONACC scores for the more general GO terms such as metabolic process (GO: anole subtree revealed conservation for several functional an- 0008152) and methylation (GO: 0032259). Finally, the A. notation features with respect to AR, suggesting that substi- carolinensis–A. apletophallus comparison yielded 42 acceler- tution rates in these genomic regions have diverged in the ated genes associated with cell adhesion (GO: 0007155), cir- anole clade with respect to the rest of the vertebrate family cadian rhythm (GO: 0007623), fertilization (GO: 0009566), tree and in the direction of stronger purifying selection. In reproductive process (GO: 0022414), response to gamma ra- contrast, the snakes subtree analysis resulted in higher con- diation (GO: 001032), and urate homeostasis (GO: 1903118). servation scores for AR and a shift toward more accelerated Out of all the pairwise-compared orthologs, putative pos- scores for intergenic regions and introns. These results suggest itively selected genes with a pairwise d /d significantly >1 N S that the evolution of each clade was accompanied by lineage- after a Bonferroni correction for multiple testing can highlight specific changes in substitution rates in functional regions, the molecular basis of adaptations unique to each species particularly stronger conservation at the DNA level for anoles. (supplementary table 15, Supplementary Material online). In We detected 3,830 genes overlapping accelerated regions A. apletophallus, these genes included GO annotations such in the anole subtree analysis, compared with 5,247 genes in as sensory perception of sound, and fertilization (sptbn4); the snake subtree analysis. Of the 1,393 genes overlapping muscle contraction (tln1); response to oxidative stress accelerated regions that are unique to anoles, 1,229 IDs were (dgkk); DNA methylation (tet3); and signal transduction and successfully mapped, resulting in enrichment of several GO protein transport (tumor suppressor candidate rhbdb2). The categories. The category with the highest fold enrichment only gene with a pairwise d /d consistent with positive N S Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 499 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE 1.00 1.00 1.00 CDS UTR5 0.75 0.75 0.75 UTR3 intergenic 0.50 0.50 0.50 introns AR 0.25 0.25 0.25 All branches Anoles Snakes 0.00 0.00 0.00 −2 0 2 −2 0 2 −2 0 2 CONACC Score CONACC Score CONACC Score FIG.4.—Distributions of conservation-acceleration (CONACC) scores for three analyses using phyloP (Pollard et al. 2010): all branches analysis including 0 0 31 vertebrates, anoles subtree analysis, and snakes subtree analysis. CDS, protein coding sequence; UTR5, 5 -untranslated region; UTR3, 3 -untranslated region;, AR, ancestral repeats. selection in A. auratus was myo18a, which is involved in met- genome-wide d /d in anoles than the other species (Kruskal– N S abolic processes and negative regulation of apoptosis. Lastly, Wallis rank sum test, supplementary fig. 6, Supplementary genes with a significantly elevated pairwise d /d in A. frena- Material online), consistent with both the pairwise d /d N S N S tus included GO terms for cellular trafficking (cog4)and cal- and the CONACC results and suggesting widespread conser- cium homeostasis (slc9a6). vation in protein-coding genes across anole lizard evolution. Pairwise d /d for genes belonging to GO terms expected a Nonetheless, using the branch-site tests we found ample ev- N S priori to be relevant to Anolis adaptive radiations were highly idence for genes under positive selection in each species that conserved and consistent with strong purifying selection are linked to relevant phenotypes important in anole lizard (fig. 5): For instance, the GO categories for pigmentation evolution. For instance, 19 genes were consistent with posi- (GO: 0043473), somitogenesis (GO: 0001756), and visual per- tive selection according to the branch-site test for A. frenatus. ception (GO: 0007601) each consisted of genes with pairwise These include plxna4, which controls cranial nerve morpho- d /d < 1 in all species comparisons. However, the distributions genesis and chemorepulsion of branchiomotor axons, affect- N S of pairwise d /d for each apriori GO category were signifi- ing the pharyngeal arches and the hyoid, which is the basis of N S cantly different (Kruskal–Wallis rank sum test), suggesting dif- the iguanian lizard dewlap (Losos 2009). Of the positively se- ferent levels of functional constraint, and within each pairwise lected genes detected in the A. auratus genome, cyp26a1 species comparison the mean d /d was highest for perception controls retinoic acid metabolism necessary for inducing Hox N S of smell (GO: 0007608), followed by either reproduction (GO: gene expression and anterior–posterior axis specification and 0000003) or limb development (GO: 0060173). These three is associated with an absent tail phenotype in the mouse— categories include genes with pairwise d /d > 1, and suggest consistent with the divergent tail morphology found in the N S accelerated evolution in pathways controlling ecologically rele- species (Irschick et al. 1997). Twenty-seven genes were found vant traits. For instance, one, 10, and 11 genes have under- to be consistent with positive selection according to the gone accelerated evolution in A. apletophallus, A. frenatus,and branch-site test for A. apletophallus, including hoxb6,which A. auratus, respectively, including the spermatogenesis- is involved in embryonic skeletal morphogenesis and anterior– associated gene ddx25 in A. frenatus. This is consistent with posterior pattern specification. In addition, we estimate that an analysis that concluded reproductive genes have evolved fzd5 in A. frenatus and ror1 in both A. frenatus and A. auratus more rapidly in the anole lineage than in other vertebrates evolved under positive selection. These genes affect beta- (Grassa and Kulathinal 2011). Another outlier with high pair- Catenin-dependent Wnt signaling, suggesting episodic posi- wise d /d was the gene cnga4 involved in perception of smell tive selection acting on different parts of the Wnt signaling N S in A. apletophallus. Finally, the analysis of limb development pathway during the evolution of anole lizards. genes in the pairwise comparison with A. frenatus revealed d /d > 1in hoxa11 and hoxd13 (see Discussion). N S Evolution of Limb Development Genes and Anole Limb After filtering for quality and complete taxon sampling, we Adaptations obtained 3,525 orthologous genes for 22 sauropsids from the MULTIZ alignment. The d /d estimates from the branch tests We found 12 limb development genes for which anoles have N S on randomized concatenated genes did not reveal a higher a significantly higher d /d than other amniotes N S 500 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 CDF CDF CDF Accelerated Evolution during Anole Lizard Diversification GBE Biological Process 2 Limb development (GO:0060173) Pigmentation (GO:0043473) Reproduction (GO:0000003) Sensory perception of smell (GO:0007608) Somitogenesis (GO:0001756) Visual perception (GO:0007601) Whole genome average Anolis frenatus Anolis auratus Anolis apletophallus FIG.5.—Distribution of d /d calculated in pairwise comparisons between the newly sequenced anole genomes and A. carolinensis for genes in adaptive N S radiation-relevant GO categories. (supplementary tables 16 and 17, Supplementary Material Supplementary Material online). In addition, compared with online). When d /d values for other amniote lineages are A. carolinensis, the hindlimbs of A. apletophallus, A. auratus, N S taken into account, the anole branches have a higher d /d and A. frenatus all exhibit elongation of the stylopod, zeugo- N S for most of the examined limb genes compared with euthe- pod, and autopod. Digit 4 is especially elongated in the hin- rian mammals (64%) and birds (68%) (fig. 6A and supple- dlimbs of all four species. Altogether, the morphological data mentary table 18, Supplementary Material online). The 12 are consistent with the molecular evolutionary results, sug- genes with higher d /d in anoles play key roles in limb de- gesting that selective pressure acting on limb phenotypes N S velopment of vertebrates (see gene expression diagrams from has driven functional changes in genes specifying the chick homologues shown in fig. 6B). For instance, hoxa10, proximal–distal axis and the hindlimb during anole lizard hoxd11,and hoxd12 displayed significantly higher d /d in diversification. N S anoles; Hox genes specify limb element identity along both the proximal–distal and cranial–caudal axes (Zuniga 2015). Discussion Interactions between FGF8 expression from the apical ecto- Here, we report a study of the genomic basis of adaptation dermal ridge (AER) and SHH in the caudal zone of polarizing during the evolution of anole lizards using complete genome activity (ZPA) leadtoprepatterningof the limbmesenchyme sequences of multiple species. Our goals were: 1) to compare and later distal limb outgrowth; hand2, grem1, bmp4,and rates of genomic divergence in anoles to other amniotes, 2) to ptch1 genes are members of this regulatory network (Zuniga understand patterns of sequence conservation in functionally 2015). The en1 gene is expressed in the ventral limb bud and annotated features in anole lizard genomes, and 3) to identify is involved in specification of the dorsal–ventral axis. Finally, putative adaptively evolving genomic regions that contributed d /d was higher in anoles for tbx5, which is required to N S to the well-studied diversification of anoles. We found that specify forelimb identity, for the transcription factor pitx1 the three de novo anole lizard genomes differ in terms of that is required for the initiation of hindlimb outgrowth, scaffold contiguity and expected gene content, and contain and for zinc finger protein sall4 that is required for cranial– largely similar transposable element profiles (although with caudal specification, particularly in the hindlimb (Infante et al. some differences—see below). We estimated a 50-Myr di- 2013; Akiyama et al. 2015; Zuniga 2015). vergence time for anoles and a background rate of neutral The morphometric analyses demonstrated that, among the molecular evolution that is greater than many other sauropsid four anole lizard species analyzed, A. frenatus was notable in groups, with strong conservation in functionally annotated having an elongated forelimb stylopod and zeugopod in re- regions compared with other vertebrates. Finally, although lation to its body size (fig. 6C and supplementary fig. 7, Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 501 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 dN/dS Tollis et al. GBE (A) (B) (C) Anolis auratus Anolis apletophallus Anolis carolinensis Anolis frenatus FIG.6.—(A) Barplot shows the differential in the estimated d /d (the parameter x)for Anolis when compared with birds and eutherian mammals, N S respectively, from a codon model incorporating d /d values for the major clades of amniotes (M2b). Asterisks indicate genes for which M2b is favored over both N S 502 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE the vast majority of protein-coding genes are highly conserved may be near extinction in Dactyloan anoles. The almost total in anole lizard genomes, we found evidence for accelerated lack of ancient repeat elements in A. carolinensis compared evolution and positive selection in genes associated with fore- with A. frenatus, A. auratus,and A. apletophallus suggests brain development, sensory perception, reproduction, and that very different evolutionary forces may have shaped the morphogenesis. We discuss these findings below. repetitive landscapes of anole lizard genomes. One explana- Our estimates of substitution rates at 4 D sites and the tion for this may be differences in each species’ demographic relative lack of conservation at ancestral repeats suggests history, which can have profound effects on genome struc- that neutral molecular evolution in anoles has been faster ture (Lynch and Conery 2003). Another explanation for than in many other sauropsid lineages, especially turtles and lineage-specific differences in transposable element diversity archosaurs. Anoles typically mate during their first year (Losos and abundance is an interaction between rates of transposi- 2009), whereas the other sequenced squamates, and even tion and DNA loss (Kapusta et al. 2017). More information more so the turtles and crocodilians, take much longer to about the recent evolutionary histories of A. frenatus, A. aur- reach sexual maturity (Tacutu et al. 2013). The results are atus,and A. apletophallus and comprehensive study of their therefore consistent with the negative relationship between repetitive DNA contents will be required to distinguish be- generation time and rates of molecular evolution that has tween which of these models of genomic repeat evolution been found across reptiles (Bromham 2002) aswellasin better explain the relative diversity, abundance, and age of mammals (Wilson Sayres et al. 2011). However, in contrast transposable elements across anole lizard genomes. to the rapid molecular evolution in neutral regions, we iden- Despite the high degree of conservation across the ge- tified strong conservation in the anole genomes at functional nome in anoles, we were able to identify numerous examples 0 0 genomic loci, based on the phyloP results for CDS, 5 -UTR, 3 - of accelerated evolution in protein-coding regions, particularly UTR and the pervasively low d /d values across protein- those associated with forebrain development and hormonal N S coding genes. Therefore, rates of molecular evolution in regulation. Anoles as a group are known to have diverse re- anoles are widely different between functional and nonfunc- productive display behaviors (Losos 2009) with sex-specific tional regions of the genome, which can be explained by rel- seasonal changes in hormonal responses (Kerver and Wade atively strong purifying selection during anole lizard 2013), and three regions of the anole forebrain have been diversification. It may be that anoles have large effective pop- linked to reproductive display behaviors, including the preoptic ulation sizes, which are less prone to fixation of alleles by ge- region, amygdala, and hypothalamus (Wade 2012). In addition, netic drift and more efficient at purging deleterious alleles. The we found evidence for positive selection on at least one gene few estimates of effective population sizes available for anoles related to the anole dewlap, plxna4 (see Results), which during suggest they may number in the hundreds of thousands or embryogenesis is expressed in the rhombomere 4 neural crest millions (Tollis and Boissinot 2014; Manthey et al. 2016), which migration group that contributes to the second pharyngeal arch is considerably greater than what has been estimated for hu- (Lumb et al. 2017). This second arch component includes the man populations, where deleterious alleles can reach high fre- hyoid apparatus that controls the dewlap (Bels 1990), which quencies (Lohmueller et al. 2008). Although some studies have features prominently in anole lizard reproductive behavior concluded more rapid turnover of deleterious retrotransposons (Wade 2012) and aids in species recognition and territorial dis- in larger populations of A. carolinensis (Tollis and Boissinot plays (Losos 2009). Rapid evolution in genes controlling the 2013; Ruggiero et al. 2017), additional population genomic development and regulation of these traits may underlie inter- studies of anoles will reveal the degree to which the fate of specific differences in sexual displays, as well as other behaviors other classes of mutation is controlled by purifying selection. that can drive prezygotic reproductive isolation and therefore Although the global repeat profiles of each anole lizard sympatric species boundaries in anoles (Ng et al. 2013). genome are largely similar, there are some notable differences The functional and behavioral importance of limb lengths in their composition, particularly in the R2/R4 non-LTR retro- in anoles is also well understood, and sympatric species occu- transposon clade, which is almost completely absent from pying different microhabitats often occupy vastly different A. frenatus. In addition, the lack of SINEs found at <5% areas of limb morphospace (Losos 2009). Morphometric K2P divergence in A. frenatus coupled with the small number analyses of the anole lizards carried out in this study revealed of copies detected in the genome suggests this element family elongation of the forelimb stylopod and zeugopod in FIG.6.—Continued M0 (the null hypothesis of a single x for the entire phylogeny) and M2a (two x model with Anolis branches as foreground); cross indicates M2a is favored over M0 and M2b using the likelihood ratio test (LRT). (B) Representations of expression pattern of limb development genes in chick embryonic limb bud, based on the Gallus Expression In Situ Hybridization Analysis (GEISHA) database (geisha.arizona.edu/geisha). (C) Replicates of Anolis auratus, A. apleto- phallus, A. carolinensis,and A. frenatus. Limb elements from the proximal stylopod (humerus or femur), middle zeugopod (radiusþ ulna or tibiaþ fibula), and distal autopod (carpusþ metacarpal IVþ digit IV phalanges or tarsusþ metatarsal IVþ digit IV phalanges) were measured in forelimb and hindlimb osteological preparations, respectively. Osteological preparations are shown to scale (2-mm bar). Additional details are in Supplementary Material online. Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 503 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE A. frenatus and elongation of all hindlimb elements in A. genomic bases of evolutionary adaptations in anole lizards frenatus, A. apletophallus,and A. auratus. Fittingly, we com- and provide great insight for vertebrate evolution in general. paredsubstitution patternsinlimbdevelopment (GO: 0060173) genes in the pairwise comparison between A. car- Supplementary Material olinensis and A. frenatus and found that hoxa11 and hoxd13, Supplementary data are available at Genome Biology and which specify limb elements along the proximal–distal and Evolution online. cranial–caudal axes, and tbx5, which specifies forelimb identity (Zuniga 2015), were outliers with pairwise d /d > 1. The out- N S growth and specification of the limb buds are regulated by Acknowledgments interactions between AER, ZPA, and limb mesenchyme involv- ing SHH and FGF regulatory network genes hand2, grem1, We thank Glenn Markov and Peter Marting for their help in bmp4,and ptch1 (Zuniga 2015), which have significantly the field and Matthew J. Huentelman, John Cornelius, higher d /d in anoles compared with other amniotes. The N S Shannon Gilpin, and Elise Kulik for technical assistance. We pitx1 and sall4 genes also have higher d /d in anoles and N S also thank Fiona McCarthy for helpful discussion. We thank are required for hindlimb bud initial outgrowth and specifica- April Allen and the Collaborative Sequencing Center at the tion, respectively (Infanteetal. 2013; Akiyamaetal. 2015; Translational Genomics Research Institute for sequencing as- Zuniga 2015). The morphometric data support elongation of sistance. Support for I.M. and E.L. was provided by the School the stylopod, zeugopod, and autopod in A. apletophallus, A. of Life Sciences Undergraduate Research (SOLUR) Program at auratus,and A. frenatus hindlimbs compared with A. caroli- Arizona State University. This work was supported by funding nensis, suggesting that adaptive evolution on phenotypes dur- from a seed grant from Arizona State University (ASU) and the ing Anolis diversification may be controlled by changes in limb Smithsonian Tropical Research Institute and funding from the development genes. Although the hindlimb gene pitx1 is rap- College of Liberal Arts and Sciences at ASU to K.K. Additional idly evolving in anoles, we did not detect accelerated evolution support to D.B.M. was provided by the National Science in tbx4 nor hoxc10, eventhough the expressionof bothgenes Foundation (CAREER IOS1149453). Computational analysis is regulated by pitx1 binding (Infante et al. 2013). Pitx1 has was supported by allocations from the Arizona State also been shown to be strongly associated with limb University Research Computing and startup from the School enhancers (Infante et al. 2013). We identified 756 brain, of Life Sciences and the Biodesign Institute to M.A.W.S. 112 liver, 109 heart, and 289 limb enhancers in A. carolinensis, which aligned to A. auratus, A. apletophallus,and A. frenatus Literature Cited with 88% through 99% identity (supplementary fig. 8 and Akiyama R, et al. 2015. Sall4-Gli3 system in early limb progenitors is es- table 19, Supplementary Material online). This high alignability sential for the development of limb skeletal elements. Proc Natl Acad across anoles of enhancers from four tissues mirrors the wide- Sci U S A. 112(16):5075–5080. spread conservation in protein coding genes, and suggests Alfo ¨ ldi J, et al. 2011. The genome of the green anole lizard and a com- that regulatory changes to limb development are modulated parative analysis with birds and mammals. Nature 477(7366):587–591. Bels VL. 1990. The mechanism of dewlap extension in Anolis carolinensis by a few specific mutations in enhancer motifs. (Reptilia: iguanidae) with histological analysis of the hyoid apparatus. J By generating de novo annotated genomes for three anole Morphol. 206(2):225–244. species, A. apletophallus, A. auratus,and A. frenatus,we Benton MJ, et al. 2015. Constraints on the timescale of animal evolution- were able to carry out a comparative genomic analysis with ary history. Palaeontol Electron 18:1–106. the A. carolinensis reference that is novel in scope for squa- Blanchette M, et al. 2004. Aligning multiple genomic sequences with the threaded blockset aligner. Genome Res. 14(4):708–715. mates. We found that the rate of molecular evolution in Blankenberg D, Taylor J, Nekrutenko A, Galaxy Team. 2011. Making anoles and snakes was very rapid relative to other vertebrates, whole genome multiple alignments usable for biologists. especially turtles and archosaurs, which may be a factor con- Bioinformatics 27(17):2426–2428. tributing to the successful adaptive radiation of anoles across Boetzer M, Henkel CV, Jansen HJ, Butler D, Pirovano W. 2011. Scaffolding the Neotropics resulting in 400 extant species. Despite this pre-assembled contigs using SSPACE. Bioinformatics 27(4):578–579. Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for rapid background rate of evolution, strong purifying selection Illumina sequence data. Bioinformatics 30(15):2114–2120. in the anole lineage has kept the vast majority of protein- Bromham L. 2002. Molecular clocks in reptiles: life history influences rate coding genes highly conserved, even relative to other verte- of molecular evolution. Mol Biol Evol. 19(3):302–309. brates. However, we did identify evidence for accelerated Bromham L, Woolfit M, Lee MSY, Rambaut A. 2002. Testing the relation- evolution in genes involved in behavior, sensory perception, ship between morphological and molecular rates of change along phylogenies. Evolution 56:1921–1930. and reproduction, as well as in genes regulating limb devel- Castoe TA, et al. 2013. The Burmese python genome reveals the molecular opment. Morphometric analyses of anole limbs corroborated basis for extreme adaptation in snakes. Proc Natl Acad Sci U S A. these findings, highlighting the importance of genes involved 110(51):20645–20650. with specification of limb outgrowth and hindlimb specifica- Consortium GO. 2015. Gene ontology consortium: going forward. Nucleic tion. Further investigations could help to elucidate the Acids Res. 43(D1):D1049–D1046. 504 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE Crawford NG, et al. 2012. More than 1000 ultraconserved elements pro- Hubisz MJ, Pollard KS, Siepel A. 2011. PHAST and RPHAST: phylogenetic vide evidence that turtles are the sister group of archosaurs. Biol Lett. analysis with space/time models. Brief Bioinformatics 12(1):41–51. 8(5):783–786. Infante CR, Park S, Mihala AG, Kingsley DM, Menke DB. 2013. Pitx1 Crawford et al. 2015. A phylogenomic analysis of turtles. Mol Phylogenet broadly associates with limb enhancers and is enriched on hindlimb Evol. 83(2015):250–257. cis-regulatory elements. Dev Biol. 374(1):234–244. Davies TJ, Savolainen V. 2006. Neutral theory, phylogenies, and the rela- Irschick DJ, Vitt LJ, Zani PA, Losos JB. 1997. A comparison of evolutionary tionship between phenotypic change and evolutionary rates. Evolution radiations in mainland and caribbean Anolis lizards. Ecology 60(3):476–483. 78(7):2191–2203. Del Rosario Castan ~eda M, De Queiroz K. 2013. Phylogeny of the Dactyloa Jarvis ED, et al. 2014. Whole-genome analyses resolve early branches in clade of Anolis lizards: new insights from combining morphological the tree of life of modern birds. Science 346(6215):1320–1331. and molecular data. Bull Mus Comp Zool. 160(7):345–398. Jones P, et al. 2014. InterProScan 5: genome-scale protein function clas- Donoghue PCJ, Benton MJ. 2007. Rocks and clocks: calibrating the Tree of sification. Bioinformatics 30(9):1236–1240. Life using fossils and molecules. Trends Ecol Evol. 22(8):424–431. Jurka J, et al. 2005. Repbase Update, a database of eukaryotic repetitive Drummond AJ, Suchard MA, Xie D, Rambaut A. 2012. Bayesian phyloge- elements. Cytogenet Genome Res. 110(1–4):462–467. netics with BEAUti and the BEAST 1.7. Mol Biol Evol. Kajitani R, et al. 2014. Efficient de novo assembly of highly heterozygous 29(8):1969–1973. genomes from whole-genome shotgun short reads. Genome Res. Eckalbar WL, et al. 2013. Genome reannotation of the lizard Anolis caro- 24(8):1384–1395. linensis based on 14 adult and embryonic deep transcriptomes. BMC Kapusta A, Suh A, Feschotte C. 2017. Dynamics of genome size evolution Genomics 14:49. in birds and mammals. Proc Natl Acad Sci U S A. 114:E1460–E1469. Edgar RC. 2004. MUSCLE: a multiple sequence alignment method with Kent WJ, Baertsch R, Hinrichs A, Miller W, Haussler D. 2003. Evolution’s reduced time and space complexity. BMC Bioinformatics 5:113. cauldron: duplication, deletion, and rearrangement in the mouse and Eldredge N, Gould J. 1972. Punctuated equilibria: an alternative to phyletic human genomes. Proc Natl Acad Sci U S A. 100(20):11484–11489. gradualism. In: Schopf TJM, editor. Models of paleobiology. San Kerver HN, Wade J. 2013. Seasonal and sexual dimorphisms in expression Francisco: Freeman, Cooper & Co. p. 82–115. of androgen receptor and its coactivators in brain and peripheral cop- Fay JC, Wu CI. 2003. Sequence divergence, functional constraint, and ulatory tissues of the green anole. Gen Comp Endocrinol. 193:56–67. selection in protein evolution. Annu Rev Genomics Hum Genet. Kim S, et al. 2016. Comparison of carnivore, omnivore, and herbivore 4:213–235. mammalian genomes with a new leopard assembly. Genome Fleishman LJ. 1988. The social behavior of Anolis auratus, a grass anole 17(1):211. from Panama. J Herpetol. 22:13. Kinsella RJ, et al. 2011. Ensembl BioMarts: a hub for data retrieval Foote AD, et al. 2015. Convergent evolution of the genomes of marine across taxonomic space. Database (Oxford) 2011(0): mammals. Nat Genet. 47(3):272–275. bar030–bar030. Genome 10K Community of Scientists 2009. Genome 10K: a proposal to Ko ¨ hler G, Sunyer J. 2008. Two new species of anoles formerly referred to obtain whole-genome sequence for 10,000 vertebrate species. J as Anolis limifrons (Squamata: polychrotidae). Herpetologica Hered. 100(6):659–674. 64(1):92–108. Georges A, et al. 2015. High-coverage sequencing and annotated assem- Korf I. 2004. Gene finding in novel genomes. BMC Bioinformatics 5:59. bly of the genome of the Australian dragon lizard Pogona vitticeps. Kumar S, Subramanian S. 2002. Mutation rates in mammalian genomes. Gigascience 4:45. Proc Natl Acad Sci U S A. 99(2):803–808. Gharib WH, Robinson-Rechavi M. 2013. The branch-site test of positive Lanfear R, Welch JJ, Bromham L. 2010. Watching the clock: studying selection is surprisingly robust but lacks power under synonymous variation in rates of molecular evolution between species. Trends substitution saturation and variation in GC. Mol Biol Evol. Ecol Evol. 25(9):495–503. 30(7):1675–1686. Lindblad-Toh K, et al. 2011. A high-resolution map of human evolutionary Gilbert C, et al. 2014. Endogenous hepadnaviruses, bornaviruses and constraint using 29 mammals. Nature 478(7370):476–482. circoviruses in snakes. Proc R Soc B. 281(1791):20141122. Liu Y, et al. 2015. Gekko japonicus genome reveals evolution of adhesive Glor RE, Losos JB, Larson A. 2005. Out of Cuba: overwater dispersal and toe pads and tail regeneration. Nat Commun. 6:10033. speciation among lizards in the Anolis carolinensis subgroup. Mol Ecol. Lohmueller KE, et al. 2008. Proportionally more deleterious genetic varia- 14(8):2419–2432. tion in European than in African populations. Nature Grassa CJ, Kulathinal RJ. 2011. Elevated evolutionary rates among func- 451(7181):994–997. tionally diverged reproductive genes across deep vertebrate lineages. Losos JB. 2009. Lizards in an evolutionary tree. Berkeley: University of Int J Evol Biol. 2011:1–9. California Press. Green RE, et al. 2014. Three crocodilian genomes reveal ancestral patterns Losos JB, Andrews RM, Sexton OJ, Schuler AL. 1991. Behavior, ecology, of evolution among archosaurs. Science 346(6215):1254449. and locomotor performance of the giant anole, Anolis frenatus. Guindon S, et al. 2010. New algorithms and methods to estimate Caribb J Sci. 27:173–179. maximum-likelihood phylogenies: assessing the performance of Losos JB, Miles DB. 2002. Testing the hypothesis that a clade has adaptively PhyML 3.0. Syst Biol. 59(3):307–321. radiated: iguanid lizard clades as a case study. Am Nat. Guyer C, Savage JM. 1986. Cladistic relationships among anoles (Sauria: 160(2):147–157. iguanidae). Syst Biol. 35(4):509–531. Lumb R, Buckberry S, Secker G, Lawrence D, Schwarz Q. 2017. Harris RS. 2007. Improved pairwise alignment of genomic DNA [Ph.D. Transcriptome profiling reveals expression signatures of cranial neural thesis]. The Pennsylvania State University. crest cells arising from different axial levels. BMC Dev Biol. 17(5):1–12. Hedges SB, Marin J, Suleski M, Paymer M, Kumar S. 2015. Tree of life Luo R, et al. 2012. SOAPdenovo2: an empirically improved memory- reveals clock-like speciation and diversification. Mol Biol Evol. efficient short-read de novo assembler. Gigascience 1(1):18. 32(4):835–845. Lynch M, Conery JS. 2003. The origins of genome complexity. Science Holt C, Yandell M. 2011. MAKER2: an annotation pipeline and genome- 302(5649):1401–1404. database management tool for second-generation genome projects. Mago c T, Salzberg SL. 2011. FLASH: fast length adjustment of short reads BMC Bioinformatics 12:491. to improve genome assemblies. Bioinformatics 27(21):2957–2963. Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 505 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE Mahler DL, Revell LJ, Glor RE, Losos JB. 2010. Ecological opportunity and completeness with single-copy orthologs. Bioinformatics 31(19): the rate of morphological evolution in the diversification of greater 3210–3212. Antillean anoles. Evolution 64(9):2731–2745. Simpson JT, et al. 2009. ABySS: a parallel assembler for short read se- Manthey JD, Tollis M, Lemmon AR, Moriarty Lemmon E, Boissinot S. 2016. quence data. Genome Res. 19(6):1117–1123. Diversification in wild populations of the model organism Anolis car- Smit AFA, Hubley RM, Green P. RepeatMasker Open-4.0 2013-2015. olinensis: a genome-wide phylogeographic investigation. Ecol Evol. http://www.repeatmasker.org. 6(22):8115–8125. Smit AFA, Hubley RM, Green P. RepeatModeler Open-1.0 2008-2015. Mitchell A, et al. 2015. The InterPro protein families database: the classifi- http://www.repeatmasker.org. cation resource after 15 years. Nucleic Acids Res. 43(D1):D213–D221. Song B, et al. 2015. A genome draft of the legless anguid lizard, Mouse Genome Sequencing Consortium, et al. 2002. Initial sequencing Ophisaurus gracilis.Gigascience 4:17. and comparative analysis of the mouse genome. Nature 420:520–562. Speir ML, et al. 2016. The UCSC Genome Browser database: 2016 update. Ng J, Landeen EL, Logsdon RM, Glor RE. 2013. Correlation between Anolis Nucleic Acids Res. 44(D1):D717–D725. lizard dewlap phenotype and environmental variation indicates adap- Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and tive divergence of a signal important to sexual selection and species post-analysis of large phylogenies. Bioinformatics 30(9):1312–1313. recognition. Evolution 67(2):573–582. Tacutu R, et al. 2013. Human Ageing Genomic Resources: integrated Nicholson KE, Crother BI, Guyer C, Savage JM. 2012. It is time for a new databases and tools for the biology and genetics of ageing. Nucleic classification of anoles (Squamata: dactyloidae). Zootaxa. 3477: Acids Res. 41(Database issue):D1027–D1033. e59741. Tarver JE, et al. 2016. The interrelationships of placental mammals and the Nicholson KE, et al. 2005. Mainland colonization by island lizards. limits of phylogenetic inference. Genome Biol Evol. 8(2):330–344. J Biogeogr. 32(6):929–938. Tollis M, Boissinot S. 2013. Lizards and LINEs: selection and demography Pagel M, Meade A, Barker D. 2004. Bayesian estimation of ancestral char- affect the fate of L1 retrotransposons in the genome of the green acter states on phylogenies. Syst Biol. 53(5):673–684. anole (Anolis carolinensis). Genome Biol Evol. 5(9):1754–1768. Pagel M, Venditti C, Meade A. 2006. Large punctuational contribution of Tollis M, Boissinot S. 2014. Genetic variation in the green anole lizard speciation to evolutionary divergence at the molecular level. Science (Anolis carolinensis) reveals island refugia and a fragmented Florida 314(5796):119–121. during the quaternary. Genetica 142(1):59–72. Parra G, Bradnam K, Ning Z, Keane T, Korf I. 2009. Assessing the gene UniProt Consortium 2015. UniProt: a hub for protein information. Nucleic space in draft genomes. Nucleic Acids Res. 37(1):289–297. Acids Res. 43:D204–D212. Pineault KM, Wellik DM. 2014. Hox genes and limb musculoskeletal de- Venditti C, Meade A, Pagel M. 2006. Detecting the node-density artifact in velopment. Curr Osteoporos Rep. 12(4):420–427. phylogeny reconstruction. Syst Biol. 55(4):637–643. Poe S. 2004. Phylogeny of anoles. Herpetol Monogr. 18:37–89. Vonk FJ, et al. 2013. The king cobra genome reveals dynamic gene evo- Pollard KS, Hubisz MJ, Rosenbloom KR, Siepel A. 2010. Detection of non- lution and adaptation in the snake venom system. Proc Natl Acad Sci U neutral substitution rates on mammalian phylogenies. Genome Res. S A. 110(51):20651–20656. 20(1):110–121. Wade J. 2012. Sculpting reproductive circuits: relationships among hor- Prates I, Rodrigues MT, Melo-Sampaio PR, Carnaval AC. 2015. mones, morphology and behavior in anole lizards. Gen Comp Phylogenetic relationships of Amazonian anole lizards (Dactyloa): tax- Endocrinol. 176(3):456–460. onomic implications, new insights about phenotypic evolution and the Wang D, Zhang Y, Zhang Z, Zhu J, Yu J. 2010. KaKs_Calculator 2.0: a timing of diversification. Mol Phylogenet Evol. 82:258–268. toolkit incorporating gamma-series methods and sliding window strat- Pyron RA, Burbrink FT, Wiens JJ. 2013. A phylogeny and revised classifi- egies. Genomics Proteomics Bioinformatics 8(1):77–80. cation of Squamata, including 4161 species of lizards and snakes. Wang Z, et al. 2013. The draft genomes of soft-shell turtle and green sea BMC Evol Biol. 13:93. turtle yield insights into the development and evolution of the turtle- Quinlan AR, Hall IM. 2010. BEDTools: a flexible suite of utilities for com- specific body plan. Nat Genet. 45(6):701–706. paring genomic features. Bioinformatics 26(6):841–842. Webster AJ, Payne RJH, Pagel M. 2003. Molecular phylogenies link rates of R Core Team. 2016. R: a language and environment for statistical com- evolution and speciation. Science 301(5632):478–478. puting. Vienna (Austria): R Foundation for Statistical Computing. Weng MP, Liao BY. 2010. MamPhEA: a web tool for mammalian pheno- Rabosky DL. 2012. Positive correlation between diversification rates and type enrichment analysis. Bioinformatics 26(17):2212–2213. phenotypic evolvability can mimic punctuated equilibrium on molec- Wiens JJ, et al. 2012. Resolving the phylogeny of lizards and snakes ular phylogenies. Evolution 66(8):2622–2627. (Squamata) with extensive sampling of genes and species. Biol Lett. Rallis C, et al. 2003. Tbx5 is required for forelimb bud formation and 8(6):1043–1046. continued outgrowth. J Embryol Exp Morphol. 130(12):2741–2751. Wilson Sayres MA, Venditti C, Pagel M, Makova KD. 2011. Do variations in Ruggiero RP, Bourgeois Y, Boissinot S. 2017. LINE insertion polymorphisms substitution rates and male mutation bias correlate with life-history traits? are abundant but at low frequencies across populations of Anolis A study of 32 mammalian genomes. Evolution 65(10):2800–2815. carolinensis. Front Genet. 8:44. Wittkopp PJ, Kalay G. 2012. Cis-regulatory elements: molecular mecha- Sanderson MJ. 2002. Estimating absolute rates of molecular evolution and nisms and evolutionary processes underlying divergence. Nature Rev divergence times: a penalized likelihood approach. Mol Biol Evol. Genet. 13(1):59–69. 19(1):101–109. Yang Z. 2007. PAML 4: phylogenetic analysis by maximum likelihood. Mol Sanderson MJ. 2003. r8s: inferring absolute rates of molecular evolution Biol Evol. 24(8):1586–1591. and divergence times in the absence of a molecular clock. Zhang G, et al. 2014. Comparative genomics reveals insights into avian Bioinformatics 19(2):301–302. genome evolution and adaptation. Science 346(6215):1311–1320. Shaffer HB, et al. 2013. The western painted turtle genome, a model for Zuniga A. 2015. Next generation limb development and evolution: old ques- the evolution of extreme physiological adaptations in a slowly evolving tions, new perspectives. J Embryol Exp Morphol. 142(22):3810–3820. lineage. Genome Biol. 14(3):R28. Simao FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. 2015. BUSCO: assessing genome assembly and annotation Associate editor: Belinda Chang 506 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Genome Biology and Evolution Oxford University Press

Loading next page...
 
/lp/ou_press/comparative-genomics-reveals-accelerated-evolution-in-conserved-bJX8b5qcmh
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution.
ISSN
1759-6653
eISSN
1759-6653
D.O.I.
10.1093/gbe/evy013
Publisher site
See Article on Publisher Site

Abstract

Squamates include all lizards and snakes, and display some of the most diverse and extreme morphological adaptations among vertebrates. However, compared with birds and mammals, relatively few resources exist for comparative genomic analyses of squamates, hampering efforts to understand the molecular bases of phenotypic diversification in such a speciose clade. In particular, the 400 species of anole lizard represent an extensive squamate radiation. Here, we sequence and assemble the draft genomes of three anole species—Anolis frenatus, Anolis auratus,and Anolis apletophallus—for com- parison with the available reference genome of Anolis carolinensis. Comparative analyses reveal a rapid background rate of molecular evolution consistent with a model of punctuated equilibrium, and strong purifying selection on functional ge- nomic elements in anoles. We find evidence for accelerated evolution in genes involved in behavior, sensory perception, and reproduction, as well as in genes regulating limb bud development and hindlimb specification. Morphometric analyses of anole fore and hindlimbs corroborated these findings. We detect signatures of positive selection across several genes related to the development and regulation of the forebrain, hormones, and the iguanian lizard dewlap, suggesting molecular changes underlying behavioral adaptations known to reinforce species boundaries were a key component in the diversifi- cation of anole lizards. Key words: Anolis, genomes, molecular evolution, substitution rates, adaptation, phenotypic evolution. The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non- commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Genome Biol. Evol. 10(2):489–506. doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 489 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE well as for unconstrained regions that are expected to accu- Introduction mulate mutations at the neutral rate. Thus, comparative ge- The field of amniote comparative genomics has benefited nomics creates the opportunity to understand the genetic from an influx of whole genome assemblies, due to efforts basis of species divergence, and untangle the links between across multiple collaborative research groups (Genome 10K molecular evolution and morphological change (Lanfear et al. Community of Scientists 2009; Lindblad-Toh et al. 2011; 2010). Green et al. 2014; Zhang et al. 2014). This has resulted in a One group of squamates with extensive diversification better phylogenetic sampling of genomes across the amniote within the Cenozoic is the anoles, which comprise almost tree of life and a vastly improved ability to understand the 400 species of mostly arboreal Neotropical lizards classified molecular mechanisms underlying the radiation of amniotes in the genus Anolis (Squamata: Dactyloidae). The morpholog- into terrestrial, aerial, freshwater, and marine habitats over ical, ecological, and behavioral disparity among anoles makes 300 Myr of evolution. For instance, the eutherian mammals this one of the most diverse squamates clades (Losos and share a time to most recent common ancestor (TMRCA) of Miles 2002), long serving as an ecological and evolutionary 100 Myr during the Cretaceous Period (Hedges et al. 2015), model with multiple examples of phenotypic convergence on and extensively diversified during the Cenozoic Era, which the islands of the Caribbean and in Central and South began 66 Ma. Each eutherian mammalian order is repre- America (Losos 2009). Anoles are an ideal group with which sented by at least one whole genome sequence (Speir et al. to test hypotheses about the relationship between rates of 2016), enabling investigations into the genomic changes lead- molecular and phenotypic evolution, as well as the genomic ingtobothdivergent (Kim et al. 2016) and convergent (Foote bases of many adaptive traits, as there is a robust reference et al. 2015) phenotypes. Similarly, modern birds also share a genome and annotation for the green anole (Anolis carolinen- Cretaceous TMRCA (Donoghue and Benton 2007; Hedges sis)(Alfo ¨ ldi et al. 2011; Eckalbaretal. 2013). However, despite et al. 2015), and the Avian Genome Consortium has begun their ongoing importance for the study of adaptive radiation to investigate the extensive avian evolution that occurred and species diversification, and existing genomic resources, throughout the Cenozoic at the genomic scale (Jarvis et al. there has not yet been a comparative genomic investigation 2014; Zhang et al. 2014). Although currently not as far along, that seeks to identify the loci underlying adaptive divergence numerous individual genome projects have been launched for during Anolis evolution. For instance, body size and relative squamate reptiles including iguanian lizards (Alfo¨ldi et al. limb lengths have strong adaptive significance in anoles, as 2011; Georges et al. 2015), snakes (Castoe et al. 2013; sympatric species interact differently with their environmental Vonk et al. 2013; Gilbert et al. 2014), anguimorphs (Song substrates and occupy separate areas of morphospace (Losos et al. 2015), and geckos (Liu et al. 2015). To better understand 2009). Although genetic control of limb development is not the mechanisms underlying phenotypic diversity in well studied in anoles, extensive work with model organisms squamates—one of the largest groups of vertebrates—and has identified potential regions of interest. For instance, Hox provide meaningful comparisons with the other major groups cluster paralogs control limb element patterning during devel- of amniotes (i.e., mammals and birds), more robust genomic opment (Pineault and Wellik 2014) and the T-box transcrip- sampling across the squamate lineages is sorely needed. tion factor 5 (tbx5) controls limb bud outgrowth (Rallis et al. Increased genomic sampling of speciose, phenotypically 2003). A comparative genomic approach to understanding diverse clades such as squamates can provide better estimates phenotypic diversity in anoles would provide a powerful of the number of mutational changes that occurred during way to scan for anole-specific mutations in candidate regions lineage diversification and the effects of those changes, if any, such as these, place these genetic changes in the context of on phenotypes. This can help determine whether or not un- what is expected under the background DNA substitution derlying shifts in molecular evolution are related to morpho- rate, and provide insights into how anoles have responded logical rates of change and species diversity, which has been a to selective pressures at the molecular level. subjectofdebate(Bromham et al. 2002; Davies and Here, we investigate molecular mechanisms of anole lizard Savolainen 2006). For instance, species are thought to diverge diversification using complete genome assemblies from four either by gradual change or by episodic bursts of rapid diver- species. Our goals were 3-fold: 1) to study patterns of geno- sification such as adaptive radiations (Eldredge and Gould mic divergence in anoles with respect to other amniote line- 1972). Punctuated episodes of speciation can cause dramatic ages, 2) to understand how the rate of genomic evolution changes in the rate in molecular evolution, possibly providing relates to functional genomic features and, in turn, pheno- a “mutational engine” for new phenotypes (Pagel et al. typic diversity in anoles, and 3) to identify putative adaptively 2006)—although there has been considerable debate over evolving genomic regions that contributed to the well-studied this issue (Lanfear et al. 2010). Genome annotations provide ecological and morphological changes that occurred during information about the boundaries between functional and the diversification of anoles. We focus our genome sequenc- nonfunctional loci, and DNA substitution rates can be esti- ing and assembly efforts on the following species with diverse mated for genomic regions exposed to natural selection as morphologies and ecological niches: 1) the Central American 490 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE giant anole (Anolis frenatus), which is a relatively large- using the Qiagen DNeasy Blood & Tissue Kit standard protocol. bodied species that resides high on tree trunks (Losos Specimen identifiers and geographic coordinates are listed in et al. 1991); 2) the grass anole (Anolis auratus), which supplementary table 1, Supplementary Material online. occupies grassy and bushy vegetation with narrow Paired-end DNA libraries were constructed with the TruSeq perches (Fleishman 1988); and 3) the slender anole Library Prep kit for target sizes of 180, 300, 500 bp, and 1 kb (Anolis apletophallus), which is primarily found lower on at the Translational Genomics Research Institute (TGen; tree trunks and on the ground (Ko ¨ hler and Sunyer 2008). Phoenix, Arizona). Mate pair libraries of target size 3 kb Combined with the reference genome for A. carolinensis, were constructed using the Nextera v1 Mate Pair Library Kit which resides in crowns and high trunks of trees in the (Illumina) at TGen and additional mate pair libraries of target southeastern United States, these species are distributed size 5 kb were constructed with the Nextera Mate Pair Library across the anole lizard phylogeny (Guyer and Savage kit at the University of Arizona Genetics Core (UAGC; Tucson, 1986; Poe 2004), which may have originated between Arizona). Genomic DNA was sequenced on both Illumina 120 and 45 Ma (Nicholson et al. 2012; Prates et al. HiSeq 2000 and 2500 platforms at TGen and UAGC, respec- 2015). The phylogenetic and ecological diversity of these tively. We trimmed raw reads for nucleotide biases, adaptors, species provides an ideal opportunity to study the geno- and a quality score cutoff of28 with Trimmomatic v0.33 mic underpinnings of Anolis diversification, adaptive radi- (Bolger et al. 2014). We corrected errors in the trimmed ations of tetrapods in general, and how evolution has sequences using the module SOAPec in SOAPdenovo2 (Luo shaped genomes and phenotypes during the history of et al. 2012). Overlapping reads were joined to form single-end land-dwelling vertebrates. reads using FLASH v1.2.8 (Magoc and Salzberg 2011). We compared the outputs of different genome assembly pipe- lines including ABySS versions 1.3.7 and 1.5.2 (Simpson Materials and Methods et al. 2009), SOAPdenovo2 (Luo et al. 2012), Platanus Animal Processing, Genome Sequencing, and Assembly v1.2.1 (Kajitani et al. 2014) for both contig and scaffold as- Adult A. apletophallus, A. auratus,and A. frenatus were col- sembly, and SSPACE v3.0 (Boetzer et al. 2011)for scaffold lected at field sites under permits SE/A-33-11 and SC/A-21-12 assembly. We selected assemblies with the longest contig and issued by the Republic of Panama Autoridad Nacional de scaffold N50 values for further analysis. Candidate assemblies Ambiente (ANAM; to K.K.) under Institutional Animal Care were also evaluated based on expected gene content using and Use Committee (IACUC) guidelines at the Smithsonian the Core Eukaryotic Gene Mapping Approach (CEGMA v2.5; Tropical Research Institute (Protocol 2011-0616-2014-07 to Parra et al. 2009), which aligns genome assemblies to a data- K.K.). Specimens were exported under ANAM permits SEX/A- base of 248 highly conserved eukaryotic proteins and reports 81-11 and SEX/A-71-12 (to K.K.) following regulation 50 CFR the completeness of detected orthologs, and Benchmarking Part 14 of the United States Fish & Wildlife Service. Lizards ~ Universal Single Copy Orthologs (BUSCO v1.2; Simao et al. were maintained under IACUC guidelines at Arizona State 2015), which scans assemblies for a vertebrate-specific data- University (Protocols 09-1053 R, 12-1274 R, and 15-1416 R; base of 3,023 conserved genes. supplementary table 1, Supplementary Material online). Specimens were euthanized according to ASU IACUC proto- Annotation and Analysis of Repeat Elements cols 10-1053 R and 12-1274 R (K.K.) via intracoelemic injec- tion of a solution of sodium pentobarbital/sodium phenytoin To analyze the repetitive landscapes of the three de novo and secondary confirmation by thoracotomy. Twenty-four anole lizard genomes, we first ran RepeatMasker v4.0.5 adult specimens were used for osteological analyses (http://www.repeatmasker.org)(Smit et al. 2013–2015) on (13 A. apletophallus;7 A. auratus;4 A. frenatus) together each assembly using a library of known Anolis repeat family with 10 adult A. carolinensis specimens obtained from consensus sequences from RepBase (Jurka et al. 2005). After Charles D. Sullivan Co., Inc. (Nashville, TN) and Marcus this initial masking of reference repeats (i.e., those known Cantos Reptiles (Fort Myers, FL). Given the lack of published from A. carolinensis and shared ancestrally across verte- studies comparing the limb morphology of A. apletophallus, brates), we scanned the unmasked portions of each genome A. auratus, A. frenatus,and A. carolinensis, we completed a for novel elements using RepeatModeler v1.0.8 (http://www. morphometric analysis of the forelimb and hindlimb stylopod repeatmasker.org)(Smit et al. 2008–2015; Smit et al. 2013– (humerus and femur), zeugopod (radius/ulna and tibia/fibula), 2015). This step allowed us to identify de novo repetitive and autopod (manus and pes) (supplementary methods and elements that may be unique to each species and thus would supplementary table 1, Supplementary Material online). In ad- not have been detected by RepeatMasker alone, which relies dition, for A. apletophallus, A. auratus,and A. frenatus, one on nucleotide similarity to previously archived elements. To specimen per species was selected for genomic study; we dis- estimate the amount of evolutionary divergence within repeat sected skeletal muscle tissue from A. apletophallus and liver families and characterize the repetitive landscape of each ge- from A. frenatus and A. auratus, and extracted genomic DNA nome, we generated repeat family-specific alignments and Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 491 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE calculated the average Kimura 2-parameter (K2P) distance sequence alignment (MSA) with MULTIZ v11.2 (Blanchette from consensus within each repeat family. We modified the et al. 2004)and used A. carolinensis as the reference species weight of two transition mutations as 1% of a single transi- and a guide phylogenetic tree (see Supplementary Material tion, in order to correct for high mutation rates at CpG sites. online). The MSA was filtered for the presence of aligned This was done for the A. auratus, A. apletophallus,and A. blocks from at least 26 out of the 31 species (84% complete). frenatus assemblies using the calcDivergenceFromAlign.pl tool contained within the RepeatMasker package. We Phylogenetic Analyses accessed the repeat landscape for A. carolinensis at http:// To reconstruct the amniote phylogeny and estimate neutral www.repeatmasker.org/species/anoCar.html (last accessed substitution rates across lineages, we relied on 4-fold degen- December 2016). erate (4 D) sites, which are positions in coding regions where all mutations are synonymous at the amino acid level and can Gene Annotations be used as a proxy of the neutral rate of evolution (Kumar and We created gene models for the A. frenatus, A. auratus,and Subramanian 2002). We extracted 4 D sites from the MSA, A. apletophallus genomes based on homology with protein- excluding sites from the mitochondrion, based on the A. car- coding sequences in A. carolinensis (AnoCar2.0) and the olinensis protein-coding sequence (CDS) annotations UniProt/SwissProt database (UniProt Consortium 2015), com- (Anocar2.0, Ensembl v86, last accessed December 2016) us- bined with ab initio predictions from SNAP (November 29, ing msa_view from the Phylogenetic Analysis with Space/Time 2013 release; Korf 2004) using multiple iterations of Models (PHAST) v1.3 package (Hubisz et al. 2011). MAKER v2.31.5 (Holt and Yandell 2011) for each species as- Phylogenetic reconstruction of the 31 included vertebrates sembly. The first iteration of MAKER aligned the protein using the 4D site data was performed with RAxML v8.2.3 sequences to the assembly to produce draft gene models. (Stamatakis 2014). We generated 20 maximum likelihood Ab initio gene predictors benefit from the training of their (ML) trees under the GTRCAT substitution model, setting Hidden Markov Models (HMM), and we trained SNAP by run- Latimeria, a lobe-finned fish, as the outgroup, and conducted ning MAKER a second time with SNAP using the species- 500 bootstrap replicates, placing the bipartitions from the specific HMMs generated from the CEGMA analysis described bootstrap analysis on the ML tree with the highest likelihood earlier (for A. apletophallus, we used the CEGMA gene mod- to assess node support. els from A. auratus, which were more complete—see Results). Using the gene models from this output, we then generated Rates of Molecular Evolution an improved HMM for SNAP in a third MAKER iteration. A We estimated absolute rates of molecular evolution in terms fourth and final run of MAKER was then performed, which of substitutions per site per million years across the best ML incorporated the final SNAP HMM and the aligned protein tree, and estimated the divergence times of the major amni- data, resulting in the final set of gene model predictions. For ote lineages, including the TMRCA of anole lizards, using the each species, the final genes were functionally annotated semiparametric method based on penalized likelihood (PL) in with two methods: 1) a BLASTp of MAKER2-predicted pro- r8s v1.8 (Sanderson 2002, 2003). We used two sets of criteria teins to known proteins in UniProt, and 2) the scanning of to constrain minimum, maximum, or fixed node ages. The MAKER-predicted proteins for the presence of functional first set of criteria was based on constraints from the paleon- domains using InterProScan 5 (Jones et al. 2014; Mitchell tological literature; specifically, the minimum age of Diapsida et al. 2015). Functional annotations were incorporated using (i.e., the TMRCA of chicken and Anolis) was set at 255.9 Myr, scripts provided with the MAKER2 package. the minimum age of Archosauria (i.e., the TMRCA of chicken and alligator) was set at 247.1 Myr, the minimum and max- Whole Genome Alignments imum ages for Mammalia (i.e., the TMRCA of platypus and As the basis for our comparative genomic study of amniotes, human) were set at 164.9–201.5 Myr, and the minimum and we generated a whole genome alignment of 31 sarcoptery- maximum ages for Neognathae (i.e., the TMRCA of chicken gian vertebrates, featuring representatives of the major and zebra finch) were set at 66–86.8 Myr (Benton et al. groups of amniotes (mammals, squamates, turtles, crocodili- 2015). The second analysis used comparatively fewer con- ans, and birds) and ten available squamate genomes including straints, using the node age estimate for Archosauria of the three de novo anole lizard genomes (supplementary table 247–250 Myr obtained from TimeTree (www.timetree.org; 2, Supplementary Material online). First, we generated pair- Hedges et al. 2015). We set the minimum and maximum wise syntenic alignments of each species’ genome to the ages for Crocodilia at 77.8–83.6 Myr (Green et al. 2014) A. carolinensis genome (AnoCar2.0) using LASTZ v1.02 and fixed the root age (i.e., the TMRCA of coelacanth and (Harris 2007), followed by chaining to form gapless blocks human) at 415 Myr in both analyses (Donoghue and Benton and netting to rank the highest scoring chains (Kent et al. 2007). The PL method estimates a different substitution rate 2003). We used the pairwise nets to construct a multiple on each branch and implements a penalty when rates differ 492 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE greatly between branches, quantified by a smoothing param- Detecting Nonneutral Substitution Rates and Identification eter for which larger values indicate a clock-like model of Anole-Accelerated Genomic Regions (Sanderson 2002). We used cross-validation to optimize In order to identify genomic regions underlying phenotypic this smoothing parameter, allowing values to range on a variation in anoles, we aimed to detect loci in our aligned log scale starting from 10 with the exponent increasing sequences that depart from neutral expectations, using 0.3 for a total of ten steps, and reran the analysis with the phyloP (Pollard et al. 2010) to estimate nonneutral substitu- optimal value. We used the gradient check implemented in tion rates both across the amniote phylogeny and specifically r8s to ensure the signs of any active constraints were cor- across the anole lineage. PhyloP computes 1) a null distribu- rect (i.e., negative if a minimum constraint was used). tion for the total number of substitutions at a locus given a As the evolution of anole lizards constitutes several radi- nonconserved (neutral) phylogenetic model, 2) an estimate of ations resulting in 400 species across two continents and the observed number of substitutions in the alignment, and 3) several islands, we predicted that estimated substitution the P value from a comparison of this estimate to the null rates in this lineage should be faster than the phylogenetic distribution calculated with a likelihood ratio test (LRT). First, average for amniotes and that this could be explained by we estimated the neutral model by fitting a time-reversible either punctuated evolution (Pagel et al. 2006) or ecolog- substitution model (REV) to the best ML phylogeny obtained ical opportunity (Mahler et al. 2010). In a punctuated sce- from the 4D site data and transforming branch lengths in nario, the total genetic distance between the root and tip terms of substitutions per site using phyloFit in PHAST. of a phylogeny (i.e., the path length) will be associated with Second, we used phyloP to produce a conservation- the number of speciation events in a lineage (Webster et al. acceleration (CONACC) score for every site in the alignment, 2003; Pagel et al. 2006). Alternatively, in the case of eco- which is a two-sided LRT-based P value converted into either a logical opportunity, the rate of evolution is correlated with negative value to indicate accelerated evolution or positive the rate of speciation (Mahler et al. 2010). In either case, value to indicate conservation. We analyzed the distributions path lengths and speciation events inferred from molecular of CONACC scores for different annotation features: CDS, data will be positively correlated (Rabosky 2012). To test 0 0 introns, 5 -UTR, 3 -UTR, intergenic regions, and ancestral this prediction, we first searched for evidence of punctu- repeats (AR), under the expectation that AR should evolve ated molecular evolution across the best ML phylogeny under relatively weaker purifying selection (Mouse Genome using a phylogenetic generalized least squares (PGLS) re- Sequencing Consortium et al. 2002) and therefore possess gression in BayesTraits v3.0 (Pagel et al. 2004), which incor- lower CONACC scores. We conducted two types of tests: porates a species phylogeny and weights data points 1) an “all branches” test that examines rate variation across according to shared ancestry. We removed the non- every branch in the phylogeny, and 2) multiple “subtree” amniote outgroup and applied a continuous model to the tests which separately examine rate variation within a partic- phylogenetic path length (i.e., the total number of substi- ular clade relative to the rest of the phylogeny. As suggested tutions per site) and the number of nodes (i.e., speciation by Pollard et al. (2010), we computed CONACC scores for events) leading to each of the 30 terminal taxa as depen- subtrees based on 10-bp windows. We compared subtree dent and independent variables, respectively. We then es- test results for anoles and snakes. As a test of the null hypoth- timated 1) the proportion of evolution across the tree esis that substitution rates between a subtree and the rest of attributed to speciation and 2) the deviation from the mo- the phylogeny are equal, we compared the subtree CONACC lecular clock due to punctuational events, using the test for score distributions for each annotation feature using the punctuational evolution available at http://www.evolution. Mann–Whitney U-test inRv3.3.0(R Core Team 2016). rdg.ac.uk/pe/index.html (last accessed April 2017), testing Finally, to determine which protein-coding genes contain for a node-density artifact which can lead to the underes- anole-accelerated regions, we used phyloP to compute one- timation of branch lengths due to regions of the phylogeny sided P values to indicate accelerated evolution (ACC). We with fewer taxa (Webster et al. 2003; Venditti et al. 2006). then collected ACC regions with a P value of 0.001 which Finally, we generated a continuous regression model using overlap with Ensembl v86 AnoCar2.0 whole gene annota- the path lengths and number of speciation events for each tions using bedtools intersect (Quinlan and Hall 2010). We taxon, and used it to predict the path lengths for A. frena- tested for statistical enrichment of overrepresented GO terms tus, A. carolinensis, A. auratus, and A. apletophallus using associated with the genes overlapping accelerated regions BayesTraits. All MCMC analyses were run for an initial using the PANTHER overrepresentation analysis tool available 10,000 generations for burn-in, followed by 1,000,000 on the Gene Ontology Consortium website (GO Ontology iterations with sampling every 1,000 generations. We database Released April 24, 2017, www.geneontology.org, used Tracer v1.6 (http://tree.bio.ed.ac.uk/software/tracer/; last accessed May 2017; Gene Ontology Consortium 2015), last accessed December 2016) to monitor MCMC conver- applying a Bonferroni correction to account for multiple hy- gence, including non auto-correlation, by ensuring an effective sample size (ESS) of 200 for all parameters. pothesis testing. Genes overlapping accelerated regions were Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 493 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE compared with the reference set of 18,527 A. carolinensis concatenated three sets of 30 randomly selected genes and genes. We also collected overrepresented GO terms from estimated d /d (the parameter x) on all terminal branches of N S ACC analyses for the snakes subtree in order to compare the sauropsid phylogeny using branch tests in codeml (mod- functional category enrichment between squamate clades. el¼ 2, Nsites¼ 0). To identify specific genes undergoing pos- itive selection in each anole lizard genome, we performed the branch-site test for each gene in codeml (model¼ 2, Selection on Protein-Coding Genes Nsites¼ 2). The gene phylogeny was inferred independently To more directly measure selective pressures acting on protein for each locus using PhyML v3.0 (Guindon et al. 2010). The coding genes during Anolis diversification and identify genes custom pipeline used to perform these analyses is available at related to ecologically relevant adaptations, we estimated the https://github.com/WilsonSayresLab/AlignmentProcessor.We ratio of nonsynonymous to synonymous substitutions (d /d ) performed independent tests with each anole lizard species as N S in a series of analyses. Genes with pairwise d /d > 1 harbor the foreground branch, comparing the likelihood of every N S an excess of nonsynonymous and potentially functionally rel- branch-site model to a null model in which x was fixed at evant changes at the amino acid level and are candidates for 1, and the likelihood ratio test (LRT) statistic (2DLnl) to the chi- inferring positive selection (Fay and Wu 2003). First, we square distribution with a degree of freedom equal to 1. downloaded Ensembl gene annotations for the A. carolinensis Genes with a P value< 0.01 after a Bonferroni adjustment genome (AnoCar2.0) from the UCSC Genome Browser (Speir for multiple testing and a x> 1 were identified as positively et al. 2016) and assembled exons by gene based on the pair- selected. When analyzing sequence data including species wise syntenic nets between A. carolinensis and A. apletophal- which diverged over 200 Ma, saturation at synonymous sites lus, A. auratus,and A. frenatus, respectively (see genome is a potential concern; however, studies have shown that the alignment methods above), using the “stitch gene blocks” branch-site test is quite robust to this issue. In fact, in this tool in Galaxy (Blankenberg et al. 2011). Orthologous gene scenario false negatives are far more likely than false positives alignments of each species to A. carolinensis were filtered for (Gharib and Robinson-Rechavi 2013), thus making our anal- quality through a custom pipeline (https://github.com/ ysis more conservative. To link positively selected genes in WilsonSayresLab/Anole_expression/tree/master/Scripts), each species to potential phenotypes, we examined statisti- which deleted frame-shift mutations and replaced internal cally enriched GO categories as above, and tested for enrich- stop codons with gaps. We calculated substitution rates at ment of mouse phenotypes using the gene IDs of the human nonsynonymous (d ) and synonymous sites (d )and esti- orthologs in MamPhEA (Weng and Liao 2010). N S mated d /d for eachde novospecies compared with A. In anoles, limb morphology is related to locomotor differ- N S carolinensis using KaKs_calculator v2.0 (Wang et al. 2010). ences and ecological niche adaptation among sympatric spe- We applied a Bonferroni adjustment for multiple hypothesis cies (Losos 2009); therefore, we predict strong selective testing to identify positively selected genes with highly signif- pressure on limb structure and function during anole lizard icant d /d > 1. To link genes with d /d > 1 to potential diversification will result in larger d /d values for limb-related N S N S N S phenotypes, we used the Ensembl gene IDs to collect GO genes. We selected 30 protein-coding genes annotated in GO terms in Biomart (Kinsella et al. 2011). We also examined processes for limb development (GO: 0060173), limb mor- the distribution of d /d for genes belonging to GO categories phogenesis (GO: 0035108), and limb bud formation (GO: N S expected apriori to be relevant to anole lizard diversification 0060174). Orthologous sequences for each gene were col- based on the literature (Losos 2009): Reproduction (GO: lected from human (Homo sapiens), mouse (Mus musculus), 0000003), pigmentation (GO: 0043473), somitogenesis cow (Bos taurus), African elephant (Loxodonta africanus), (GO: 0001756), visual perception (GO: 0007601), perception American opossum (Monodelphis domesticus), western of smell (GO: 0007608), and limb development (GO: clawed frog (Xenopus tropicalis), chicken (Gallus gallus), 0060173) (Gene Ontology Consortium 2015). wild turkey (Meleagris gallopavo), mallard duck (Anas platyr- To analyze substitution rates in protein-coding genes hyncos), zebra finch (Taenopygia guttata), medium ground across sauropsids and identify positively selected genes in finch (Geospiza fortis), budgerigar (Melopsittacus undulatus), the four anole lizard genomes, we tested codon-based mod- painted turtle (Chrysemys picta), Chinese softshell turtle els of molecular evolution with codeml in PAML v4.9 (Yang (Pelodiscus sinensis), American alligator (Alligator mississip- 2007). We filtered the MSA to include only the 22 sauropsids piensis), Burmese python (Python bivittatus), and green anole (including representative squamates, crocodilians, birds, and (Anolis carolinensis). For the three de novo anole lizard turtles), stitched exons for each gene with Galaxy based on genomes, we generated predicted transcript sequence using the Ensembl v86 AnoCar2.0 annotations and quality filtered genomic scaffold sequences aligning to A. carolinensis exons for frame shifts and stop codons as above, and retained only by BLASTn and genome annotation data. Homologous gene gene alignments for which all species were present. To deter- sequences were aligned by codon using MUSCLE (Edgar mine if genome-wide d /d in protein-coding genes is related 2004) and edited by eye to remove ambiguously aligning N S to the background rate of molecular evolution in anoles, we regions and stop codons, with care taken to maintain the 494 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE integrity of the amino acid sequence by frequent BLAST A. frenatus, A. auratus,and A. apletophallus, respectively. All searches to NCBI. We analyzed the codon alignments of three de novo draft assemblies were 2Gb inlength, compa- limb genes using branch tests in codeml. For each gene, we rablein sizeto the estimated 2.2-Gb A. carolinensis genome assumed the species tree topology and compared three mod- and 1.78-Gb AnoCar2.0 genome assembly (Alfo ¨ ldi et al. els which allow x to vary along different branches in the 2011). Genome assembly statistics are summarized in table 2. phylogeny: 1) the null model (M0; model¼ 0, Nsites¼ 0) in The four anole genome assemblies differed in terms of which a single x is estimated for all branches on the phylog- estimated gene content, with the assemblies of A. frenatus eny, 2) an alternative model (M2a; model¼ 2, Nsites¼ 0) that and A. auratus approaching similar completeness to that of allows one x for the Anolis branches and another for all other A. carolinensis (table 2 and fig. 1A). In the A. frenatus assem- branches, and 3) a second alternative model (M2b; mod- bly, 86% of core eukaryotic genes (CEGs) were at least par- el¼ 2, Nsites¼ 0) that allows x to vary among Anolis, python, tially represented, whereas A. auratus had 97% at least alligator, turtles, birds, eutherian mammals, and all remaining partially represented CEGs; and both are similar to A. caroli- branches. We compared each alternative model to the null nensis (AnoCar2.0), which contains 91% of CEGs. In the case model by comparing the LRT statistic to the chi square distri- of A. apletophallus, 65% of CEGs were represented. BUSCO bution given the degrees of freedom determined by the dif- analyses estimate slightly lower gene representation than ference in the number of model parameters. Where the null CEGMA across all anole genomes, including 71% for was rejected, we then compared the two alternative models A. frenatus,85% for A. auratus, and 47% for A. apletophal- using the LRT. To compare relative selective pressure on limb lus, compared with 88% BUSCO representation in A. caroli- development genes among taxa, we observed the plus or nensis. The relatively low estimated gene content in the minus differential in x between Anolis and eutherian mam- A. apletophallus assembly likely reflects high fragmentation mals and birds, respectively. due to the lower sequencing coverage and high heterozygos- ity in the sequenced A. apletophallus specimen. Indeed, anal- ysis of the kmer frequency distribution generated from the Conservation of Regulatory Elements A. apletophallus Illumina reads was highly bimodal (supple- Much of the phenotypic diversity within and between species mentary fig. 1, Supplementary Material online) and suggestive is due to cis-regulatory sequences that control gene expres- of a high proportion of heterozygous sites, which can hinder sion, including enhancers (Wittkopp and Kalay 2012). In order genome assembly efforts (Kajitani et al. 2014). Analysis of the to determine if mutations in enhancer elements affected evo- distribution of GC content across the three de novo anole lutionary change during the evolution of anole lizards, we lizard genomes reveals similar distributions which lie within annotated mouse enhancer regions identified from brain, the peaks of both A. carolinensis and chicken (Gallus gallus, liver, limb, and heart tissues and compared their conservation fig. 1B), albeit with potentially more GC-bias in A. frenatus. in A. carolinensis, A. auratus, A. apletophallus,and A. frenatus Using both homology-based and ab initio methods for the (supplementary methods, Supplementary Material online). final gene annotations for each de novo anole lizard assembly, we approached the number of predicted protein-coding genes in the A. carolinensis genome (22,962 genes predicted; Results Eckalbar et al. 2013, table 2). Large proportions of annotated Genome Assemblies and Gene Annotations genes consist of BLASTp matches to the SwissProt-UniProtKB We carried out Illumina platform-based whole genome se- database and contain known protein domains; thus, all three quencing of three anole species representing both Dactyloa de novo anole genomes presented here contain genes with and Norops clades, using combinations of high coverage clear orthology in other species. The A. auratus assembly con- paired-end sequencing plus “jumping” mate pair libraries of tains 19,838 annotated protein-coding genes, with 87% varying sizes (table 1). We sequenced the Dactyloa group (17,210 genes) having significant BLASTp matches, and species A. frenatus at 116 coverage, and the two Norops 88% (17,394 genes) with known protein domains. We an- group species, A. auratus and A. apletophallus, at 141 cov- notated 19,923 protein-coding genes in the A. frenatus as- erage and 56 coverage, respectively. The lower coverage for sembly; 89% (17,771 genes) of these contain a significant A. apletophallus is attributable to technical failures in the se- BLASTp match and 92% (18,353 genes) contain protein quencing of some of the 300-bp insert libraries, the down- domains. Meanwhile, we were able to annotate 12,816 stream effects of which we attempted to ameliorate by genes for A. apletophallus; 82% of these produced both sig- adding additional coverage using 500-bp insert libraries. nificant BLASTp hits (10,483 genes) and protein domains Anolis frenatus and A. auratus contigs were assembled with (10,534 genes). ABySS v1.3.7 and scaffolded with SOAPdenovo2; A. apleto- Repetitive elements make up similar proportions of each phallus contigs and scaffolds were assembled using ABySS sequenced Anolis lizard genome (A. carolinensis¼ 33%, (supplementary tables 3–5, Supplementary Material online). A. frenatus¼ 37%, A. auratus¼ 39%, A. apletophallus¼ 29%; Scaffold N50 statistics calculated were 37, 49, and 9.5 kb for supplementary tables 6–9, Supplementary Material online). Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 495 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE Table 1 Whole Genome Sequencing Data Obtained from Three Anole Lizard Species Libraries Anolis frenatus Anolis auratus Anolis apletophallus Total Data (Mb) Coverage Total Data (Mb) Coverage Total Data (Mb) Coverage 200-bp paired-end 1,400 70 1,058 43 359 17 300-bp paired-end 502 22 1,454 63 328 11 500-bp paired-end – – – – 154 7 1-kb paired-end 94 9 515 23 244 11 3-kb mate pair 628 11 452 8 404 7 5-kb mate pair 90 4 81 4 85 4 Total 2,714 116 3,560 141 1,574 56 Table 2 Descriptive Statistics for Four Anole Lizard Genome Assemblies Included in This Study Anolis carolinensis (AnoCar2.0) Anolis frenatus (Afre1.0) Anolis auratus (Aaur1.0) Anolis apletophallus (Aapl1.0) Contigs Scaffolds Contigs Scaffolds Contigs Scaffolds Contigs Scaffolds N50 (bp) 79,867 150,641,573 23,240 36,910 19,858 48,995 2,534 9,520 L50 (Number) 6,217 5 21,942 14,499 26,569 11,107 206,073 53,667 Longest (bp) 582,046 263,920,458 448,330 649,467 304,182 563,800 110,998 217,008 Total length (Gb) 1.70 1.79 1.92 2.03 1.94 2.02 1.89 2.18 Number 1 kb 41,987 6,457 176,895 649,467 230,878 141,345 659,833 454,194 Number 100 kb 4,247 852 809 2,134 392 3,038 1 103 Percent gaps 5% 6% 4% 14% BUSCOs C: 73%[D: 1.4%], C: 50%[D: 0.7%], C: 68%,[D: 1.1%], C: 28%[D: 2.1%], F: 15%, M: 12% F: 20%, M: 29% F: 16%, M: 15% F: 19%, M: 53% Number of 22,962 19,923 19,838 12,816 annotated genes Percent repeats 33% 37% 39% 29% NOTE.—BUSCO, Benchmarking Single Copy Orthologs; C, Complete; D, Duplicated; F, Fragmented; M, Missing. We found evidence for the differential accumulation and signatures of ancient L1 and L2 retrotransposon expan- loss of transposable elements across anole lizard genomes, sions. The lack of SINEs found at <5% K2P divergence in such as R2 non-LTR retrotransposons which occupy be- A. frenatus, coupled with the low number of copies we tween 0.92% and 1.37% of the genomes of A. carolinen- detected in the genome, suggest this element family may sis, A. auratus, and A. apletophallus genomes yet are be near extinction in the Dactyloan lineage of anoles. scarcely present in A. frenatus at 0.04%. Other differences lie in the abundance of SINEs, which are three times more Phylogenetic Results and Support for a Cenozoic Origin of abundant in A. carolinensis than in A. frenatus. The starkest Anoles difference between the repetitive landscapes of the four Anolis genomes is the evidence of recent activity coupled We reconstructed the ML phylogeny of 31 sarcopterygian with the loss of ancient elements in A. carolinensis, which vertebrates, including four anole lizards, using a DNA align- stands out in comparison with the other three assemblies ment of 4 D sites that contained 763,131 distinct patterns (supplementary fig. 2, Supplementary Material online). For with 5.17% gaps or undetermined characters. All branches instance, the divergence profile of interspersed repeats in of the highest scoring ML tree had full support from the boot- A. carolinensis is characterized by a relative lack of trans- strap replicates, and the topology was highly congruent with posable elements beyond 5–10% K2P divergence from other phylogenomic studies of amniotes (fig. 2 and their family consensus and is indicative of their recent in- supplementary fig. 3 and table 10, Supplementary Material sertion in the genome. In contrast, the A. auratus, A. aple- online). This includes several previously controversial nodes in tophallus, and A. frenatus assemblies harbor a greater the amniote phylogeny, most notably a monophyletic proportion of their elements far beyond this level of diver- Atlantogenata (including elephant and armadillo) in the mam- gence (i.e., up to 45% K2P) and contain, for example, the malian subtree (Tarver et al. 2016) and full support for a 496 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE (A) (B) 0.15 Complete Fragmented Anolis carolinensis Anolis frenatus BUSCO CEGMA Anolis auratus ● Anolis apletophallus 0.10 Gallus gallus 0.05 ● ●●● ●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●● 0.00 20 40 60 80 GC Percentage FIG.1.—(A) Proportion of conserved orthologs in either complete or fragmented form detected in four anole lizard genomes. Proportions are reported out of 3,023 BUSCO (Benchmarking Universal Single Copy Orthologs) and 248 CEGMA (Conserved Eukaryotic Gene Mapping Approach) orthologs. (B) Distribution of GC content in the genomes of four anole lizards and chicken, calculated in 5-kb windows. turtle-archosaur clade (Archelosauria) (Crawford et al. 2012, (Glor et al. 2005), and the Central American A. auratus and 2015; Green et al. 2014). Within the included squamates, the A. apletophallus was estimated to be 35 Myr, suggesting that gecko forms the basal lineage, snakes are monophyletic and the recolonization of the mainland by Caribbean anoles of the the four anole lizards and Pogona support a monophyletic Norops lineage (Nicholson et al. 2005) occurred after the late Iguania, consistent with several studies (Wiens et al. 2012; Eocene. The r8s analysis using fewer constraints resulted in Pyron et al. 2013). We obtained full support for the some slightly different node age estimates, particularly in Dactyloan A. frenatus belonging to the basal lineage of mammals; however the TMRCA for anoles did not differ sig- anoles, followed by the divergence of A. carolinensis and nificantly. We also used a read-mapping approach to recon- monophyletic Norops species A. auratus and A. apletophallus, struct the mitochondrial genomes of A. frenatus and consistent with other phylogenetic studies of anoles (Guyer A. auratus, obtained orthologous mitogenomic data from and Savage 1986; Nicholson et al. 2012; Del Rosario iguanian lizards available on NCBI (supplementary table 11, Castaneda and De Queiroz 2013; Prates et al. 2015). Supplementary Material Online), and used an internal fossil The fossil-based node constraints we implemented in r8s calibration to estimate divergence times with a partitioned passed all gradient checks, and our estimates for the origin of data set using BEAST v1.8 (Drummond et al. 2012)(supple- amniotes at 303 Ma, the origin of sauropsids at 278 Ma, and mentary methods, Supplementary Material Online). This mi- the origin of placental mammals at 99 Ma are consistent with tochondrial analysis was done prior to obtaining the sequence previously published estimates (Donoghue and Benton 2007; data for A. apletophallus and thus this species was not in- supplementary table 10, Supplementary Material online). cluded. The results from the BEAST analysis were consistent Regarding the lineage leading to squamates including anoles, with those from the 4D site analysis, placing the 95% highest we estimated that the gecko lineage split from other squa- posterior density (HPD) of the TMRCA for iguanian lizards mates 175 Ma, and the TMRCA for iguanian lizards was between 195 and 113 Myr, the TMRCA of A. frenatus and 136 Myr. The TMRCA of the anoles was estimated to be A. carolinensis between 67 and 41 Myr (mean 53 Myr), and 45 Myr, suggesting a Cenozoic origin, specifically during the the TMRCA of A. carolinensis and A. auratus between 52 and Eocene which lasted between 56 and 34 Ma. The TMRCA of 27 Myr (mean 40 Myr) (supplementary fig. 4, Supplementary Anolis carolinensis, which is native to southeastern Material Online). Thus, both 4 D sites and mitogenomic data North America and originated from a Cuban radiation support a hypothesis in which anole lizard evolution occurred Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 497 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Anolis carolinensis Anolis frenatus Anolis auratus Anolis apletophallus Anolis carolinensis Anolis frenatus Anolis auratus Anolis apletophallus Proportion of Conserved Orthologs Detected Proportion of 5kb Regions Tollis et al. GBE Anolis apletophallus (A) Anolis auratus 0.0020 Anolis carolinensis Anolis frenatus Pogona vitticeps Python bivittatus Boa constricor 0.0018 Ophiophagus hannah Crotalus mitchellii Gekko japonicus Pelodiscus sinensis Chrysemys picta bellii 0.0016 ● Alligator mississippiensis Crocodylus porosus Gavialis gangeticus Struthio camelus 0.0014 Anas platyrhynchos Meleagris gallopavo Gallus gallus Melopsittacus undulatus Taenopygia guttata Geospiza fortis Ornithorhynchus anatinus Monodelphis domestica (B) Dasypus novemcinctus Loxodonta africana Bos taurus Mus musculus Canis lupus familiaris Mus musculus Homo sapiens 0.8 Latimeria chalumnae ● Anolis apletophallus Anolis auratus ●● ●● ●● 0.2 ● ● ● Anolis frenatus FIG.2.—Phylogeny of 31 vertebrates from 763,131 4-fold degenerate Anolis carolinensis ● 0.6 sites reconstructed using maximum likelihood in RAxML v8.2.3 (Stamatakis 2014). All branches received full support from 500 bootstrap replicates. Representative images of species are not drawn to scale. ● 0.4 within the Cenozoic Era, and that the Norops colonization of ● ●● Central and South America began after the late Eocene. ● Chrysemys picta bellii Speciation Events Accelerated Molecular Evolution in Anoles A comparison of the branch lengths of the phylogeny in FIG.3.—(A) Comparison of the estimated DNA substitution rates (in terms of substitutions per site per million years) on terminal and internal figure 2 suggests rapid evolution at 4 D sites along the squa- branches of the amniote phylogeny. (B) Scatterplot showing the positive mate branches, including the ancestral anole branch, as well relationship (R ¼ 0.17, P¼ 0.02) between the total phylogenetic path as the common ancestor branches of birds and mammals, length (i.e., the sum of substitutions per site from the root to the terminal respectively. Using the PL method, estimated rates of neutral taxon) and the number of speciation events (i.e., the number of nodes divergence at 4 D sites varied greatly across amniotes, and leading to the terminal taxon). Values for the four anole lizards studied differed significantly among phylogenetic groups (fig. 3A). here and two outlier taxa (house mouse, Mus musculus and western The optimal value for the smoothing parameter was 1, sug- painted turtle, Chrysemys picta bellii) are shown for comparison. Color gesting significant deviation from the molecular clock. We legend is as follows: mammals¼ orange, squamates¼ green, tur- observed that, among the sauropsids included in this study, tles¼ blue, crocodilians¼ violet, birds¼ gray. anoles and snakes have diverged at the fastest evolutionary since the split from their shared archosaurian ancestor rates in terms of DNA substitutions per site per million years. 247 Ma. Finally, DNA substitution rates at 4 D sites in mam- Consistent with previous studies, we estimated that the turtle mals surpass those of all other amniotes. lineage evolved at a far slower rate (Shaffer et al. 2013; Wang We found support for punctuated evolution across et al. 2013), as has the crocodilian lineage (Green et al. 2014). amniotes at 4 D sites, with the slope of the regression be- We also found that birds have departed significantly from tween path lengths and speciation events significantly >0 crocodilians in terms of DNA substitution rates at 4 D sites 498 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Anoles Snakes Other Squamates Turtles Crocodilians Birds Mammals Phylogenetic Path Length DNA Substitution Rate Accelerated Evolution during Anole Lizard Diversification GBE (R ¼ 0.171, P< 0.02, 95% HPD¼ 0.170–0.174, fig. 3B). In was diencephalon development (GO: 0021536, 4.37-fold en- addition, 26% of evolution across the amniote phylogeny was richment, P¼ 4.26 10 ), which was represented by 14 attributed to speciation events, with 17% of deviation from genes (vs. 3.2 expected), followed by endocrine system devel- the molecular clock caused by punctuational evolution. We opment (GO: 0035270, 3.69-fold enrichment, P¼ 4.20 found no evidence for the node-density artifact. Consistent 10 ), whichwas representedby 17genes (vs. 4.60 with the PL analysis, we estimated from the regression that the expected). We detected 2,810 genes overlapping accelerated turtle, crocodilian, and avian lineages have evolved at compar- regions unique to snakes; GO terms with the highest fold atively much slower rates that are below the trend line for enrichment in snakes were artery morphogenesis (GO: punctuated evolution. In contrast, mammals and squamates 0048844, 3-fold enrichment, 27 observed vs. 9 expected, were above this trend line. Convergence was reached in all P¼ 5.78 10 ) and artery development (GO: 0035904, 3- MCMC analyses, with ESS for all parameters 200. The re- fold enrichment, 23 observed vs. 7.8 expected, gression model accurately predicted the estimated path P¼ 4.51 10 ). lengths for A. frenatus (median 0.69 substitutions per site both predicted and observed, 95% HPD 0.50–0.90), A. caro- Accelerated Evolution and Positive Selection Acting on linensis (median 0.72 predicted vs. 0.71 observed, 95% HPD Protein-Coding Genes during Anolis Diversification 0.53–0.93), A. auratus (median 0.75 predicted vs. 0.73 ob- served, 95% HPD 0.53–0.96), and A. apletophallus (median We computed d /d through pairwise comparisons of N S 0.76 predicted vs. 0.73 observed, 95% HPD 0.57–0.99). 10,991, 13,783, and 12,246 filtered orthologous genes be- These results suggest that rapid molecular evolution in anoles tween A. carolinensis and A. frenatus, A. auratus,and is linked to rapid speciation rates during lineage diversification. A. apletophallus, respectively. Overall, we found high levels of protein-coding sequence conservation, with 88–93% of pairwise d /d estimates for genes in each comparison <0.5 N S Pervasive Conservation at the DNA Level for Functional (supplementary tables 12–14, Supplementary Material on- Elements in Anoles line), suggesting that the evolution for a majority of genes We used CONACC scores from phyloP to assess departures can be characterized by strong purifying selection (supple- 0 0 from neutrality in CDS, 5 -UTR, 3 -UTR, introns, intergenic mentary fig. 5, Supplementary Material online). Pairwise d / regions, and AR, first across all branches of the vertebrate d was >1 for 53 genes in the A. carolinensis–A. frenatus phylogeny and then in subtree analyses of anoles and snakes comparison, including genes associated with specific GO (fig. 4). The distribution of CONACC scores showed clear terms for biological processes such as locomotion (GO: differences by genomic feature in the all branches test. For 0040011), cellular response to insulin-like growth factor stim- instance, CDS sequences were by far the most conserved of ulus (GO: 1990314), execution phase of apoptosis (GO: the analyzed functional elements, with the highest levels of 0097194), and regulation of reactive oxygen species (GO: acceleration in AR, consistent with expectations that evolution 1903426), possibly reflecting adaptive differences in adult in coding regions is constrained with respect to nonfunctional body size, metabolic rate, and locomotor performance be- repetitive regions. In both subtree analyses, CONACC scores tween the two species. The 30 accelerated genes in the for AR were significantly different from other annotation fea- A. carolinensis–A. auratus comparison were associated with tures (Mann–Whitney U-test, P¼ 0). CONACC scores for the more general GO terms such as metabolic process (GO: anole subtree revealed conservation for several functional an- 0008152) and methylation (GO: 0032259). Finally, the A. notation features with respect to AR, suggesting that substi- carolinensis–A. apletophallus comparison yielded 42 acceler- tution rates in these genomic regions have diverged in the ated genes associated with cell adhesion (GO: 0007155), cir- anole clade with respect to the rest of the vertebrate family cadian rhythm (GO: 0007623), fertilization (GO: 0009566), tree and in the direction of stronger purifying selection. In reproductive process (GO: 0022414), response to gamma ra- contrast, the snakes subtree analysis resulted in higher con- diation (GO: 001032), and urate homeostasis (GO: 1903118). servation scores for AR and a shift toward more accelerated Out of all the pairwise-compared orthologs, putative pos- scores for intergenic regions and introns. These results suggest itively selected genes with a pairwise d /d significantly >1 N S that the evolution of each clade was accompanied by lineage- after a Bonferroni correction for multiple testing can highlight specific changes in substitution rates in functional regions, the molecular basis of adaptations unique to each species particularly stronger conservation at the DNA level for anoles. (supplementary table 15, Supplementary Material online). In We detected 3,830 genes overlapping accelerated regions A. apletophallus, these genes included GO annotations such in the anole subtree analysis, compared with 5,247 genes in as sensory perception of sound, and fertilization (sptbn4); the snake subtree analysis. Of the 1,393 genes overlapping muscle contraction (tln1); response to oxidative stress accelerated regions that are unique to anoles, 1,229 IDs were (dgkk); DNA methylation (tet3); and signal transduction and successfully mapped, resulting in enrichment of several GO protein transport (tumor suppressor candidate rhbdb2). The categories. The category with the highest fold enrichment only gene with a pairwise d /d consistent with positive N S Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 499 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE 1.00 1.00 1.00 CDS UTR5 0.75 0.75 0.75 UTR3 intergenic 0.50 0.50 0.50 introns AR 0.25 0.25 0.25 All branches Anoles Snakes 0.00 0.00 0.00 −2 0 2 −2 0 2 −2 0 2 CONACC Score CONACC Score CONACC Score FIG.4.—Distributions of conservation-acceleration (CONACC) scores for three analyses using phyloP (Pollard et al. 2010): all branches analysis including 0 0 31 vertebrates, anoles subtree analysis, and snakes subtree analysis. CDS, protein coding sequence; UTR5, 5 -untranslated region; UTR3, 3 -untranslated region;, AR, ancestral repeats. selection in A. auratus was myo18a, which is involved in met- genome-wide d /d in anoles than the other species (Kruskal– N S abolic processes and negative regulation of apoptosis. Lastly, Wallis rank sum test, supplementary fig. 6, Supplementary genes with a significantly elevated pairwise d /d in A. frena- Material online), consistent with both the pairwise d /d N S N S tus included GO terms for cellular trafficking (cog4)and cal- and the CONACC results and suggesting widespread conser- cium homeostasis (slc9a6). vation in protein-coding genes across anole lizard evolution. Pairwise d /d for genes belonging to GO terms expected a Nonetheless, using the branch-site tests we found ample ev- N S priori to be relevant to Anolis adaptive radiations were highly idence for genes under positive selection in each species that conserved and consistent with strong purifying selection are linked to relevant phenotypes important in anole lizard (fig. 5): For instance, the GO categories for pigmentation evolution. For instance, 19 genes were consistent with posi- (GO: 0043473), somitogenesis (GO: 0001756), and visual per- tive selection according to the branch-site test for A. frenatus. ception (GO: 0007601) each consisted of genes with pairwise These include plxna4, which controls cranial nerve morpho- d /d < 1 in all species comparisons. However, the distributions genesis and chemorepulsion of branchiomotor axons, affect- N S of pairwise d /d for each apriori GO category were signifi- ing the pharyngeal arches and the hyoid, which is the basis of N S cantly different (Kruskal–Wallis rank sum test), suggesting dif- the iguanian lizard dewlap (Losos 2009). Of the positively se- ferent levels of functional constraint, and within each pairwise lected genes detected in the A. auratus genome, cyp26a1 species comparison the mean d /d was highest for perception controls retinoic acid metabolism necessary for inducing Hox N S of smell (GO: 0007608), followed by either reproduction (GO: gene expression and anterior–posterior axis specification and 0000003) or limb development (GO: 0060173). These three is associated with an absent tail phenotype in the mouse— categories include genes with pairwise d /d > 1, and suggest consistent with the divergent tail morphology found in the N S accelerated evolution in pathways controlling ecologically rele- species (Irschick et al. 1997). Twenty-seven genes were found vant traits. For instance, one, 10, and 11 genes have under- to be consistent with positive selection according to the gone accelerated evolution in A. apletophallus, A. frenatus,and branch-site test for A. apletophallus, including hoxb6,which A. auratus, respectively, including the spermatogenesis- is involved in embryonic skeletal morphogenesis and anterior– associated gene ddx25 in A. frenatus. This is consistent with posterior pattern specification. In addition, we estimate that an analysis that concluded reproductive genes have evolved fzd5 in A. frenatus and ror1 in both A. frenatus and A. auratus more rapidly in the anole lineage than in other vertebrates evolved under positive selection. These genes affect beta- (Grassa and Kulathinal 2011). Another outlier with high pair- Catenin-dependent Wnt signaling, suggesting episodic posi- wise d /d was the gene cnga4 involved in perception of smell tive selection acting on different parts of the Wnt signaling N S in A. apletophallus. Finally, the analysis of limb development pathway during the evolution of anole lizards. genes in the pairwise comparison with A. frenatus revealed d /d > 1in hoxa11 and hoxd13 (see Discussion). N S Evolution of Limb Development Genes and Anole Limb After filtering for quality and complete taxon sampling, we Adaptations obtained 3,525 orthologous genes for 22 sauropsids from the MULTIZ alignment. The d /d estimates from the branch tests We found 12 limb development genes for which anoles have N S on randomized concatenated genes did not reveal a higher a significantly higher d /d than other amniotes N S 500 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 CDF CDF CDF Accelerated Evolution during Anole Lizard Diversification GBE Biological Process 2 Limb development (GO:0060173) Pigmentation (GO:0043473) Reproduction (GO:0000003) Sensory perception of smell (GO:0007608) Somitogenesis (GO:0001756) Visual perception (GO:0007601) Whole genome average Anolis frenatus Anolis auratus Anolis apletophallus FIG.5.—Distribution of d /d calculated in pairwise comparisons between the newly sequenced anole genomes and A. carolinensis for genes in adaptive N S radiation-relevant GO categories. (supplementary tables 16 and 17, Supplementary Material Supplementary Material online). In addition, compared with online). When d /d values for other amniote lineages are A. carolinensis, the hindlimbs of A. apletophallus, A. auratus, N S taken into account, the anole branches have a higher d /d and A. frenatus all exhibit elongation of the stylopod, zeugo- N S for most of the examined limb genes compared with euthe- pod, and autopod. Digit 4 is especially elongated in the hin- rian mammals (64%) and birds (68%) (fig. 6A and supple- dlimbs of all four species. Altogether, the morphological data mentary table 18, Supplementary Material online). The 12 are consistent with the molecular evolutionary results, sug- genes with higher d /d in anoles play key roles in limb de- gesting that selective pressure acting on limb phenotypes N S velopment of vertebrates (see gene expression diagrams from has driven functional changes in genes specifying the chick homologues shown in fig. 6B). For instance, hoxa10, proximal–distal axis and the hindlimb during anole lizard hoxd11,and hoxd12 displayed significantly higher d /d in diversification. N S anoles; Hox genes specify limb element identity along both the proximal–distal and cranial–caudal axes (Zuniga 2015). Discussion Interactions between FGF8 expression from the apical ecto- Here, we report a study of the genomic basis of adaptation dermal ridge (AER) and SHH in the caudal zone of polarizing during the evolution of anole lizards using complete genome activity (ZPA) leadtoprepatterningof the limbmesenchyme sequences of multiple species. Our goals were: 1) to compare and later distal limb outgrowth; hand2, grem1, bmp4,and rates of genomic divergence in anoles to other amniotes, 2) to ptch1 genes are members of this regulatory network (Zuniga understand patterns of sequence conservation in functionally 2015). The en1 gene is expressed in the ventral limb bud and annotated features in anole lizard genomes, and 3) to identify is involved in specification of the dorsal–ventral axis. Finally, putative adaptively evolving genomic regions that contributed d /d was higher in anoles for tbx5, which is required to N S to the well-studied diversification of anoles. We found that specify forelimb identity, for the transcription factor pitx1 the three de novo anole lizard genomes differ in terms of that is required for the initiation of hindlimb outgrowth, scaffold contiguity and expected gene content, and contain and for zinc finger protein sall4 that is required for cranial– largely similar transposable element profiles (although with caudal specification, particularly in the hindlimb (Infante et al. some differences—see below). We estimated a 50-Myr di- 2013; Akiyama et al. 2015; Zuniga 2015). vergence time for anoles and a background rate of neutral The morphometric analyses demonstrated that, among the molecular evolution that is greater than many other sauropsid four anole lizard species analyzed, A. frenatus was notable in groups, with strong conservation in functionally annotated having an elongated forelimb stylopod and zeugopod in re- regions compared with other vertebrates. Finally, although lation to its body size (fig. 6C and supplementary fig. 7, Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 501 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 dN/dS Tollis et al. GBE (A) (B) (C) Anolis auratus Anolis apletophallus Anolis carolinensis Anolis frenatus FIG.6.—(A) Barplot shows the differential in the estimated d /d (the parameter x)for Anolis when compared with birds and eutherian mammals, N S respectively, from a codon model incorporating d /d values for the major clades of amniotes (M2b). Asterisks indicate genes for which M2b is favored over both N S 502 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE the vast majority of protein-coding genes are highly conserved may be near extinction in Dactyloan anoles. The almost total in anole lizard genomes, we found evidence for accelerated lack of ancient repeat elements in A. carolinensis compared evolution and positive selection in genes associated with fore- with A. frenatus, A. auratus,and A. apletophallus suggests brain development, sensory perception, reproduction, and that very different evolutionary forces may have shaped the morphogenesis. We discuss these findings below. repetitive landscapes of anole lizard genomes. One explana- Our estimates of substitution rates at 4 D sites and the tion for this may be differences in each species’ demographic relative lack of conservation at ancestral repeats suggests history, which can have profound effects on genome struc- that neutral molecular evolution in anoles has been faster ture (Lynch and Conery 2003). Another explanation for than in many other sauropsid lineages, especially turtles and lineage-specific differences in transposable element diversity archosaurs. Anoles typically mate during their first year (Losos and abundance is an interaction between rates of transposi- 2009), whereas the other sequenced squamates, and even tion and DNA loss (Kapusta et al. 2017). More information more so the turtles and crocodilians, take much longer to about the recent evolutionary histories of A. frenatus, A. aur- reach sexual maturity (Tacutu et al. 2013). The results are atus,and A. apletophallus and comprehensive study of their therefore consistent with the negative relationship between repetitive DNA contents will be required to distinguish be- generation time and rates of molecular evolution that has tween which of these models of genomic repeat evolution been found across reptiles (Bromham 2002) aswellasin better explain the relative diversity, abundance, and age of mammals (Wilson Sayres et al. 2011). However, in contrast transposable elements across anole lizard genomes. to the rapid molecular evolution in neutral regions, we iden- Despite the high degree of conservation across the ge- tified strong conservation in the anole genomes at functional nome in anoles, we were able to identify numerous examples 0 0 genomic loci, based on the phyloP results for CDS, 5 -UTR, 3 - of accelerated evolution in protein-coding regions, particularly UTR and the pervasively low d /d values across protein- those associated with forebrain development and hormonal N S coding genes. Therefore, rates of molecular evolution in regulation. Anoles as a group are known to have diverse re- anoles are widely different between functional and nonfunc- productive display behaviors (Losos 2009) with sex-specific tional regions of the genome, which can be explained by rel- seasonal changes in hormonal responses (Kerver and Wade atively strong purifying selection during anole lizard 2013), and three regions of the anole forebrain have been diversification. It may be that anoles have large effective pop- linked to reproductive display behaviors, including the preoptic ulation sizes, which are less prone to fixation of alleles by ge- region, amygdala, and hypothalamus (Wade 2012). In addition, netic drift and more efficient at purging deleterious alleles. The we found evidence for positive selection on at least one gene few estimates of effective population sizes available for anoles related to the anole dewlap, plxna4 (see Results), which during suggest they may number in the hundreds of thousands or embryogenesis is expressed in the rhombomere 4 neural crest millions (Tollis and Boissinot 2014; Manthey et al. 2016), which migration group that contributes to the second pharyngeal arch is considerably greater than what has been estimated for hu- (Lumb et al. 2017). This second arch component includes the man populations, where deleterious alleles can reach high fre- hyoid apparatus that controls the dewlap (Bels 1990), which quencies (Lohmueller et al. 2008). Although some studies have features prominently in anole lizard reproductive behavior concluded more rapid turnover of deleterious retrotransposons (Wade 2012) and aids in species recognition and territorial dis- in larger populations of A. carolinensis (Tollis and Boissinot plays (Losos 2009). Rapid evolution in genes controlling the 2013; Ruggiero et al. 2017), additional population genomic development and regulation of these traits may underlie inter- studies of anoles will reveal the degree to which the fate of specific differences in sexual displays, as well as other behaviors other classes of mutation is controlled by purifying selection. that can drive prezygotic reproductive isolation and therefore Although the global repeat profiles of each anole lizard sympatric species boundaries in anoles (Ng et al. 2013). genome are largely similar, there are some notable differences The functional and behavioral importance of limb lengths in their composition, particularly in the R2/R4 non-LTR retro- in anoles is also well understood, and sympatric species occu- transposon clade, which is almost completely absent from pying different microhabitats often occupy vastly different A. frenatus. In addition, the lack of SINEs found at <5% areas of limb morphospace (Losos 2009). Morphometric K2P divergence in A. frenatus coupled with the small number analyses of the anole lizards carried out in this study revealed of copies detected in the genome suggests this element family elongation of the forelimb stylopod and zeugopod in FIG.6.—Continued M0 (the null hypothesis of a single x for the entire phylogeny) and M2a (two x model with Anolis branches as foreground); cross indicates M2a is favored over M0 and M2b using the likelihood ratio test (LRT). (B) Representations of expression pattern of limb development genes in chick embryonic limb bud, based on the Gallus Expression In Situ Hybridization Analysis (GEISHA) database (geisha.arizona.edu/geisha). (C) Replicates of Anolis auratus, A. apleto- phallus, A. carolinensis,and A. frenatus. Limb elements from the proximal stylopod (humerus or femur), middle zeugopod (radiusþ ulna or tibiaþ fibula), and distal autopod (carpusþ metacarpal IVþ digit IV phalanges or tarsusþ metatarsal IVþ digit IV phalanges) were measured in forelimb and hindlimb osteological preparations, respectively. Osteological preparations are shown to scale (2-mm bar). Additional details are in Supplementary Material online. Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 503 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE A. frenatus and elongation of all hindlimb elements in A. genomic bases of evolutionary adaptations in anole lizards frenatus, A. apletophallus,and A. auratus. Fittingly, we com- and provide great insight for vertebrate evolution in general. paredsubstitution patternsinlimbdevelopment (GO: 0060173) genes in the pairwise comparison between A. car- Supplementary Material olinensis and A. frenatus and found that hoxa11 and hoxd13, Supplementary data are available at Genome Biology and which specify limb elements along the proximal–distal and Evolution online. cranial–caudal axes, and tbx5, which specifies forelimb identity (Zuniga 2015), were outliers with pairwise d /d > 1. The out- N S growth and specification of the limb buds are regulated by Acknowledgments interactions between AER, ZPA, and limb mesenchyme involv- ing SHH and FGF regulatory network genes hand2, grem1, We thank Glenn Markov and Peter Marting for their help in bmp4,and ptch1 (Zuniga 2015), which have significantly the field and Matthew J. Huentelman, John Cornelius, higher d /d in anoles compared with other amniotes. The N S Shannon Gilpin, and Elise Kulik for technical assistance. We pitx1 and sall4 genes also have higher d /d in anoles and N S also thank Fiona McCarthy for helpful discussion. We thank are required for hindlimb bud initial outgrowth and specifica- April Allen and the Collaborative Sequencing Center at the tion, respectively (Infanteetal. 2013; Akiyamaetal. 2015; Translational Genomics Research Institute for sequencing as- Zuniga 2015). The morphometric data support elongation of sistance. Support for I.M. and E.L. was provided by the School the stylopod, zeugopod, and autopod in A. apletophallus, A. of Life Sciences Undergraduate Research (SOLUR) Program at auratus,and A. frenatus hindlimbs compared with A. caroli- Arizona State University. This work was supported by funding nensis, suggesting that adaptive evolution on phenotypes dur- from a seed grant from Arizona State University (ASU) and the ing Anolis diversification may be controlled by changes in limb Smithsonian Tropical Research Institute and funding from the development genes. Although the hindlimb gene pitx1 is rap- College of Liberal Arts and Sciences at ASU to K.K. Additional idly evolving in anoles, we did not detect accelerated evolution support to D.B.M. was provided by the National Science in tbx4 nor hoxc10, eventhough the expressionof bothgenes Foundation (CAREER IOS1149453). Computational analysis is regulated by pitx1 binding (Infante et al. 2013). Pitx1 has was supported by allocations from the Arizona State also been shown to be strongly associated with limb University Research Computing and startup from the School enhancers (Infante et al. 2013). We identified 756 brain, of Life Sciences and the Biodesign Institute to M.A.W.S. 112 liver, 109 heart, and 289 limb enhancers in A. carolinensis, which aligned to A. auratus, A. apletophallus,and A. frenatus Literature Cited with 88% through 99% identity (supplementary fig. 8 and Akiyama R, et al. 2015. Sall4-Gli3 system in early limb progenitors is es- table 19, Supplementary Material online). This high alignability sential for the development of limb skeletal elements. Proc Natl Acad across anoles of enhancers from four tissues mirrors the wide- Sci U S A. 112(16):5075–5080. spread conservation in protein coding genes, and suggests Alfo ¨ ldi J, et al. 2011. The genome of the green anole lizard and a com- that regulatory changes to limb development are modulated parative analysis with birds and mammals. Nature 477(7366):587–591. Bels VL. 1990. The mechanism of dewlap extension in Anolis carolinensis by a few specific mutations in enhancer motifs. (Reptilia: iguanidae) with histological analysis of the hyoid apparatus. J By generating de novo annotated genomes for three anole Morphol. 206(2):225–244. species, A. apletophallus, A. auratus,and A. frenatus,we Benton MJ, et al. 2015. Constraints on the timescale of animal evolution- were able to carry out a comparative genomic analysis with ary history. Palaeontol Electron 18:1–106. the A. carolinensis reference that is novel in scope for squa- Blanchette M, et al. 2004. Aligning multiple genomic sequences with the threaded blockset aligner. Genome Res. 14(4):708–715. mates. We found that the rate of molecular evolution in Blankenberg D, Taylor J, Nekrutenko A, Galaxy Team. 2011. Making anoles and snakes was very rapid relative to other vertebrates, whole genome multiple alignments usable for biologists. especially turtles and archosaurs, which may be a factor con- Bioinformatics 27(17):2426–2428. tributing to the successful adaptive radiation of anoles across Boetzer M, Henkel CV, Jansen HJ, Butler D, Pirovano W. 2011. Scaffolding the Neotropics resulting in 400 extant species. Despite this pre-assembled contigs using SSPACE. Bioinformatics 27(4):578–579. Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for rapid background rate of evolution, strong purifying selection Illumina sequence data. Bioinformatics 30(15):2114–2120. in the anole lineage has kept the vast majority of protein- Bromham L. 2002. Molecular clocks in reptiles: life history influences rate coding genes highly conserved, even relative to other verte- of molecular evolution. Mol Biol Evol. 19(3):302–309. brates. However, we did identify evidence for accelerated Bromham L, Woolfit M, Lee MSY, Rambaut A. 2002. Testing the relation- evolution in genes involved in behavior, sensory perception, ship between morphological and molecular rates of change along phylogenies. Evolution 56:1921–1930. and reproduction, as well as in genes regulating limb devel- Castoe TA, et al. 2013. The Burmese python genome reveals the molecular opment. Morphometric analyses of anole limbs corroborated basis for extreme adaptation in snakes. Proc Natl Acad Sci U S A. these findings, highlighting the importance of genes involved 110(51):20645–20650. with specification of limb outgrowth and hindlimb specifica- Consortium GO. 2015. Gene ontology consortium: going forward. Nucleic tion. Further investigations could help to elucidate the Acids Res. 43(D1):D1049–D1046. 504 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Accelerated Evolution during Anole Lizard Diversification GBE Crawford NG, et al. 2012. More than 1000 ultraconserved elements pro- Hubisz MJ, Pollard KS, Siepel A. 2011. PHAST and RPHAST: phylogenetic vide evidence that turtles are the sister group of archosaurs. Biol Lett. analysis with space/time models. Brief Bioinformatics 12(1):41–51. 8(5):783–786. Infante CR, Park S, Mihala AG, Kingsley DM, Menke DB. 2013. Pitx1 Crawford et al. 2015. A phylogenomic analysis of turtles. Mol Phylogenet broadly associates with limb enhancers and is enriched on hindlimb Evol. 83(2015):250–257. cis-regulatory elements. Dev Biol. 374(1):234–244. Davies TJ, Savolainen V. 2006. Neutral theory, phylogenies, and the rela- Irschick DJ, Vitt LJ, Zani PA, Losos JB. 1997. A comparison of evolutionary tionship between phenotypic change and evolutionary rates. Evolution radiations in mainland and caribbean Anolis lizards. Ecology 60(3):476–483. 78(7):2191–2203. Del Rosario Castan ~eda M, De Queiroz K. 2013. Phylogeny of the Dactyloa Jarvis ED, et al. 2014. Whole-genome analyses resolve early branches in clade of Anolis lizards: new insights from combining morphological the tree of life of modern birds. Science 346(6215):1320–1331. and molecular data. Bull Mus Comp Zool. 160(7):345–398. Jones P, et al. 2014. InterProScan 5: genome-scale protein function clas- Donoghue PCJ, Benton MJ. 2007. Rocks and clocks: calibrating the Tree of sification. Bioinformatics 30(9):1236–1240. Life using fossils and molecules. Trends Ecol Evol. 22(8):424–431. Jurka J, et al. 2005. Repbase Update, a database of eukaryotic repetitive Drummond AJ, Suchard MA, Xie D, Rambaut A. 2012. Bayesian phyloge- elements. Cytogenet Genome Res. 110(1–4):462–467. netics with BEAUti and the BEAST 1.7. Mol Biol Evol. Kajitani R, et al. 2014. Efficient de novo assembly of highly heterozygous 29(8):1969–1973. genomes from whole-genome shotgun short reads. Genome Res. Eckalbar WL, et al. 2013. Genome reannotation of the lizard Anolis caro- 24(8):1384–1395. linensis based on 14 adult and embryonic deep transcriptomes. BMC Kapusta A, Suh A, Feschotte C. 2017. Dynamics of genome size evolution Genomics 14:49. in birds and mammals. Proc Natl Acad Sci U S A. 114:E1460–E1469. Edgar RC. 2004. MUSCLE: a multiple sequence alignment method with Kent WJ, Baertsch R, Hinrichs A, Miller W, Haussler D. 2003. Evolution’s reduced time and space complexity. BMC Bioinformatics 5:113. cauldron: duplication, deletion, and rearrangement in the mouse and Eldredge N, Gould J. 1972. Punctuated equilibria: an alternative to phyletic human genomes. Proc Natl Acad Sci U S A. 100(20):11484–11489. gradualism. In: Schopf TJM, editor. Models of paleobiology. San Kerver HN, Wade J. 2013. Seasonal and sexual dimorphisms in expression Francisco: Freeman, Cooper & Co. p. 82–115. of androgen receptor and its coactivators in brain and peripheral cop- Fay JC, Wu CI. 2003. Sequence divergence, functional constraint, and ulatory tissues of the green anole. Gen Comp Endocrinol. 193:56–67. selection in protein evolution. Annu Rev Genomics Hum Genet. Kim S, et al. 2016. Comparison of carnivore, omnivore, and herbivore 4:213–235. mammalian genomes with a new leopard assembly. Genome Fleishman LJ. 1988. The social behavior of Anolis auratus, a grass anole 17(1):211. from Panama. J Herpetol. 22:13. Kinsella RJ, et al. 2011. Ensembl BioMarts: a hub for data retrieval Foote AD, et al. 2015. Convergent evolution of the genomes of marine across taxonomic space. Database (Oxford) 2011(0): mammals. Nat Genet. 47(3):272–275. bar030–bar030. Genome 10K Community of Scientists 2009. Genome 10K: a proposal to Ko ¨ hler G, Sunyer J. 2008. Two new species of anoles formerly referred to obtain whole-genome sequence for 10,000 vertebrate species. J as Anolis limifrons (Squamata: polychrotidae). Herpetologica Hered. 100(6):659–674. 64(1):92–108. Georges A, et al. 2015. High-coverage sequencing and annotated assem- Korf I. 2004. Gene finding in novel genomes. BMC Bioinformatics 5:59. bly of the genome of the Australian dragon lizard Pogona vitticeps. Kumar S, Subramanian S. 2002. Mutation rates in mammalian genomes. Gigascience 4:45. Proc Natl Acad Sci U S A. 99(2):803–808. Gharib WH, Robinson-Rechavi M. 2013. The branch-site test of positive Lanfear R, Welch JJ, Bromham L. 2010. Watching the clock: studying selection is surprisingly robust but lacks power under synonymous variation in rates of molecular evolution between species. Trends substitution saturation and variation in GC. Mol Biol Evol. Ecol Evol. 25(9):495–503. 30(7):1675–1686. Lindblad-Toh K, et al. 2011. A high-resolution map of human evolutionary Gilbert C, et al. 2014. Endogenous hepadnaviruses, bornaviruses and constraint using 29 mammals. Nature 478(7370):476–482. circoviruses in snakes. Proc R Soc B. 281(1791):20141122. Liu Y, et al. 2015. Gekko japonicus genome reveals evolution of adhesive Glor RE, Losos JB, Larson A. 2005. Out of Cuba: overwater dispersal and toe pads and tail regeneration. Nat Commun. 6:10033. speciation among lizards in the Anolis carolinensis subgroup. Mol Ecol. Lohmueller KE, et al. 2008. Proportionally more deleterious genetic varia- 14(8):2419–2432. tion in European than in African populations. Nature Grassa CJ, Kulathinal RJ. 2011. Elevated evolutionary rates among func- 451(7181):994–997. tionally diverged reproductive genes across deep vertebrate lineages. Losos JB. 2009. Lizards in an evolutionary tree. Berkeley: University of Int J Evol Biol. 2011:1–9. California Press. Green RE, et al. 2014. Three crocodilian genomes reveal ancestral patterns Losos JB, Andrews RM, Sexton OJ, Schuler AL. 1991. Behavior, ecology, of evolution among archosaurs. Science 346(6215):1254449. and locomotor performance of the giant anole, Anolis frenatus. Guindon S, et al. 2010. New algorithms and methods to estimate Caribb J Sci. 27:173–179. maximum-likelihood phylogenies: assessing the performance of Losos JB, Miles DB. 2002. Testing the hypothesis that a clade has adaptively PhyML 3.0. Syst Biol. 59(3):307–321. radiated: iguanid lizard clades as a case study. Am Nat. Guyer C, Savage JM. 1986. Cladistic relationships among anoles (Sauria: 160(2):147–157. iguanidae). Syst Biol. 35(4):509–531. Lumb R, Buckberry S, Secker G, Lawrence D, Schwarz Q. 2017. Harris RS. 2007. Improved pairwise alignment of genomic DNA [Ph.D. Transcriptome profiling reveals expression signatures of cranial neural thesis]. The Pennsylvania State University. crest cells arising from different axial levels. BMC Dev Biol. 17(5):1–12. Hedges SB, Marin J, Suleski M, Paymer M, Kumar S. 2015. Tree of life Luo R, et al. 2012. SOAPdenovo2: an empirically improved memory- reveals clock-like speciation and diversification. Mol Biol Evol. efficient short-read de novo assembler. Gigascience 1(1):18. 32(4):835–845. Lynch M, Conery JS. 2003. The origins of genome complexity. Science Holt C, Yandell M. 2011. MAKER2: an annotation pipeline and genome- 302(5649):1401–1404. database management tool for second-generation genome projects. Mago c T, Salzberg SL. 2011. FLASH: fast length adjustment of short reads BMC Bioinformatics 12:491. to improve genome assemblies. Bioinformatics 27(21):2957–2963. Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 505 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Tollis et al. GBE Mahler DL, Revell LJ, Glor RE, Losos JB. 2010. Ecological opportunity and completeness with single-copy orthologs. Bioinformatics 31(19): the rate of morphological evolution in the diversification of greater 3210–3212. Antillean anoles. Evolution 64(9):2731–2745. Simpson JT, et al. 2009. ABySS: a parallel assembler for short read se- Manthey JD, Tollis M, Lemmon AR, Moriarty Lemmon E, Boissinot S. 2016. quence data. Genome Res. 19(6):1117–1123. Diversification in wild populations of the model organism Anolis car- Smit AFA, Hubley RM, Green P. RepeatMasker Open-4.0 2013-2015. olinensis: a genome-wide phylogeographic investigation. Ecol Evol. http://www.repeatmasker.org. 6(22):8115–8125. Smit AFA, Hubley RM, Green P. RepeatModeler Open-1.0 2008-2015. Mitchell A, et al. 2015. The InterPro protein families database: the classifi- http://www.repeatmasker.org. cation resource after 15 years. Nucleic Acids Res. 43(D1):D213–D221. Song B, et al. 2015. A genome draft of the legless anguid lizard, Mouse Genome Sequencing Consortium, et al. 2002. Initial sequencing Ophisaurus gracilis.Gigascience 4:17. and comparative analysis of the mouse genome. Nature 420:520–562. Speir ML, et al. 2016. The UCSC Genome Browser database: 2016 update. Ng J, Landeen EL, Logsdon RM, Glor RE. 2013. Correlation between Anolis Nucleic Acids Res. 44(D1):D717–D725. lizard dewlap phenotype and environmental variation indicates adap- Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and tive divergence of a signal important to sexual selection and species post-analysis of large phylogenies. Bioinformatics 30(9):1312–1313. recognition. Evolution 67(2):573–582. Tacutu R, et al. 2013. Human Ageing Genomic Resources: integrated Nicholson KE, Crother BI, Guyer C, Savage JM. 2012. It is time for a new databases and tools for the biology and genetics of ageing. Nucleic classification of anoles (Squamata: dactyloidae). Zootaxa. 3477: Acids Res. 41(Database issue):D1027–D1033. e59741. Tarver JE, et al. 2016. The interrelationships of placental mammals and the Nicholson KE, et al. 2005. Mainland colonization by island lizards. limits of phylogenetic inference. Genome Biol Evol. 8(2):330–344. J Biogeogr. 32(6):929–938. Tollis M, Boissinot S. 2013. Lizards and LINEs: selection and demography Pagel M, Meade A, Barker D. 2004. Bayesian estimation of ancestral char- affect the fate of L1 retrotransposons in the genome of the green acter states on phylogenies. Syst Biol. 53(5):673–684. anole (Anolis carolinensis). Genome Biol Evol. 5(9):1754–1768. Pagel M, Venditti C, Meade A. 2006. Large punctuational contribution of Tollis M, Boissinot S. 2014. Genetic variation in the green anole lizard speciation to evolutionary divergence at the molecular level. Science (Anolis carolinensis) reveals island refugia and a fragmented Florida 314(5796):119–121. during the quaternary. Genetica 142(1):59–72. Parra G, Bradnam K, Ning Z, Keane T, Korf I. 2009. Assessing the gene UniProt Consortium 2015. UniProt: a hub for protein information. Nucleic space in draft genomes. Nucleic Acids Res. 37(1):289–297. Acids Res. 43:D204–D212. Pineault KM, Wellik DM. 2014. Hox genes and limb musculoskeletal de- Venditti C, Meade A, Pagel M. 2006. Detecting the node-density artifact in velopment. Curr Osteoporos Rep. 12(4):420–427. phylogeny reconstruction. Syst Biol. 55(4):637–643. Poe S. 2004. Phylogeny of anoles. Herpetol Monogr. 18:37–89. Vonk FJ, et al. 2013. The king cobra genome reveals dynamic gene evo- Pollard KS, Hubisz MJ, Rosenbloom KR, Siepel A. 2010. Detection of non- lution and adaptation in the snake venom system. Proc Natl Acad Sci U neutral substitution rates on mammalian phylogenies. Genome Res. S A. 110(51):20651–20656. 20(1):110–121. Wade J. 2012. Sculpting reproductive circuits: relationships among hor- Prates I, Rodrigues MT, Melo-Sampaio PR, Carnaval AC. 2015. mones, morphology and behavior in anole lizards. Gen Comp Phylogenetic relationships of Amazonian anole lizards (Dactyloa): tax- Endocrinol. 176(3):456–460. onomic implications, new insights about phenotypic evolution and the Wang D, Zhang Y, Zhang Z, Zhu J, Yu J. 2010. KaKs_Calculator 2.0: a timing of diversification. Mol Phylogenet Evol. 82:258–268. toolkit incorporating gamma-series methods and sliding window strat- Pyron RA, Burbrink FT, Wiens JJ. 2013. A phylogeny and revised classifi- egies. Genomics Proteomics Bioinformatics 8(1):77–80. cation of Squamata, including 4161 species of lizards and snakes. Wang Z, et al. 2013. The draft genomes of soft-shell turtle and green sea BMC Evol Biol. 13:93. turtle yield insights into the development and evolution of the turtle- Quinlan AR, Hall IM. 2010. BEDTools: a flexible suite of utilities for com- specific body plan. Nat Genet. 45(6):701–706. paring genomic features. Bioinformatics 26(6):841–842. Webster AJ, Payne RJH, Pagel M. 2003. Molecular phylogenies link rates of R Core Team. 2016. R: a language and environment for statistical com- evolution and speciation. Science 301(5632):478–478. puting. Vienna (Austria): R Foundation for Statistical Computing. Weng MP, Liao BY. 2010. MamPhEA: a web tool for mammalian pheno- Rabosky DL. 2012. Positive correlation between diversification rates and type enrichment analysis. Bioinformatics 26(17):2212–2213. phenotypic evolvability can mimic punctuated equilibrium on molec- Wiens JJ, et al. 2012. Resolving the phylogeny of lizards and snakes ular phylogenies. Evolution 66(8):2622–2627. (Squamata) with extensive sampling of genes and species. Biol Lett. Rallis C, et al. 2003. Tbx5 is required for forelimb bud formation and 8(6):1043–1046. continued outgrowth. J Embryol Exp Morphol. 130(12):2741–2751. Wilson Sayres MA, Venditti C, Pagel M, Makova KD. 2011. Do variations in Ruggiero RP, Bourgeois Y, Boissinot S. 2017. LINE insertion polymorphisms substitution rates and male mutation bias correlate with life-history traits? are abundant but at low frequencies across populations of Anolis A study of 32 mammalian genomes. Evolution 65(10):2800–2815. carolinensis. Front Genet. 8:44. Wittkopp PJ, Kalay G. 2012. Cis-regulatory elements: molecular mecha- Sanderson MJ. 2002. Estimating absolute rates of molecular evolution and nisms and evolutionary processes underlying divergence. Nature Rev divergence times: a penalized likelihood approach. Mol Biol Evol. Genet. 13(1):59–69. 19(1):101–109. Yang Z. 2007. PAML 4: phylogenetic analysis by maximum likelihood. Mol Sanderson MJ. 2003. r8s: inferring absolute rates of molecular evolution Biol Evol. 24(8):1586–1591. and divergence times in the absence of a molecular clock. Zhang G, et al. 2014. Comparative genomics reveals insights into avian Bioinformatics 19(2):301–302. genome evolution and adaptation. Science 346(6215):1311–1320. Shaffer HB, et al. 2013. The western painted turtle genome, a model for Zuniga A. 2015. Next generation limb development and evolution: old ques- the evolution of extreme physiological adaptations in a slowly evolving tions, new perspectives. J Embryol Exp Morphol. 142(22):3810–3820. lineage. Genome Biol. 14(3):R28. Simao FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. 2015. BUSCO: assessing genome assembly and annotation Associate editor: Belinda Chang 506 Genome Biol. Evol. 10(2):489–506 doi:10.1093/gbe/evy013 Advance Access publication January 19, 2018 Downloaded from https://academic.oup.com/gbe/article-abstract/10/2/489/4817506 by Ed 'DeepDyve' Gillespie user on 16 March 2018

Journal

Genome Biology and EvolutionOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from Google Scholar, PubMed
Create lists to organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off